Tokenizer: ./smlm_pretrained_iter5_0/tokenizer Model: ./smlm_pretrained_iter5_0/model
	Train size: 80 Test size: 20


		-------------RUN 1-----------
			------------EPOCH 1---------------
Loss:  tensor(2580.6885, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2350.4788, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2818.4663, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1224.0269, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1947.0482, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1723.9897, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1084.3499, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1345.8243, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1222.2371, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1226.6761, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2072.4351, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2762.3789, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1211.7332, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1855.9136, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2486.4016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3896.5776, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2550.0876, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2658.0029, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2230.3931, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1755.7080, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2012.6108, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1402.6837, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2181.9219, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1779.1616, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2769.4512, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2418.2993, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2471.0752, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1235.9510, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2306.4351, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2262.4883, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1267.8524, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1802.1455, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2301.4023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1510.1919, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2180.6152, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1903.4175, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1974.3223, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1893.8010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1824.7488, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2070.5730, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2313.7168, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1889.4131, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2980.7583, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1203.9192, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1752.9987, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1773.1851, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1665.0898, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1508.5406, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1644.1688, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1063.6700, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1872.6414, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2768.2280, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2797.9722, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2088.5859, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1228.9321, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1393.9011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1592.5828, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1449.8647, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.058823529411764705, 'recall': 0.011857707509881422, 'f1': 0.019736842105263157, 'number': 253}, 'P': {'precision': 0.01839080459770115, 'recall': 0.023323615160349854, 'f1': 0.020565552699228794, 'number': 343}, 'overall_precision': 0.02263374485596708, 'overall_recall': 0.018456375838926176, 'overall_f1': 0.02033271719038817, 'overall_accuracy': 0.49056868382301383}
			------------EPOCH 2---------------
Loss:  tensor(1893.7817, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1775.2316, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1977.9753, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(901.1757, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1510.4556, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1259.3252, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(790.8087, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(998.0347, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1019.8455, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(975.3680, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1700.4561, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2188.3477, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1056.8403, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1478.5977, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2082.6414, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3324.1401, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2163.6958, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2266.4644, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1950.2998, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1432.6309, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1520.4905, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1157.7664, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1720.2478, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1584.0626, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2509.0745, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2082.0923, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2288.5820, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1069.4545, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1770.8950, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1816.8839, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1076.0778, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1586.0952, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1887.0049, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1182.1232, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1716.1885, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1543.8308, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1651.2444, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1479.4558, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1368.5032, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1692.8562, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1965.3157, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1437.4272, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2549.5623, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(952.3057, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1393.8975, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1428.8042, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1415.2136, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1310.8707, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1407.0343, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(912.2598, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1439.7142, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2308.5554, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2474.0159, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1815.2043, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(982.3018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1147.8450, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1291.1794, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1215.2847, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.0989010989010989, 'recall': 0.03557312252964427, 'f1': 0.05232558139534884, 'number': 253}, 'P': {'precision': 0.0847457627118644, 'recall': 0.13119533527696792, 'f1': 0.10297482837528604, 'number': 343}, 'overall_precision': 0.08681672025723473, 'overall_recall': 0.09060402684563758, 'overall_f1': 0.08866995073891626, 'overall_accuracy': 0.5941378584951967}
			------------EPOCH 3---------------
Loss:  tensor(1456.3499, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1451.0055, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1573.9933, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(689.8297, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1196.5931, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1006.3893, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(647.7760, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(825.1262, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(824.5510, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(750.5764, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1406.5900, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1807.0635, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(910.6877, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1154.0767, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1809.1952, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2910.0588, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1817.6003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1955.9968, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1644.0276, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1113.4774, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1212.5046, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(903.1403, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1396.0873, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1412.2219, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1959.7655, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1713.5261, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1952.0524, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(789.0110, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1308.4834, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1398.4509, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(892.0482, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1315.8313, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1474.6475, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(813.8661, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1397.0159, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1137.0942, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1330.7338, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1013.8691, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(881.4231, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1251.2140, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1702.0143, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1009.3230, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1922.1223, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(730.5112, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1070.0404, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1184.8595, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1175.1167, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1094.0583, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1159.3577, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(748.8844, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1040.0876, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1906.2761, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1852.7458, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1541.5549, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(717.6498, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(898.1585, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1006.8416, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(981.2043, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.13392857142857142, 'recall': 0.11857707509881422, 'f1': 0.12578616352201258, 'number': 253}, 'P': {'precision': 0.1595959595959596, 'recall': 0.2303206997084548, 'f1': 0.18854415274463007, 'number': 343}, 'overall_precision': 0.15159944367176634, 'overall_recall': 0.18288590604026847, 'overall_f1': 0.16577946768060836, 'overall_accuracy': 0.6383142837108197}
			------------EPOCH 4---------------
Loss:  tensor(1026.8243, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1042.3420, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1196.1204, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(496.5488, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(897.9188, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(717.1315, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(500.7084, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(615.4738, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(687.3188, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(540.1472, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1205.3906, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1552.0991, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(718.5737, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(917.6653, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1641.0549, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2549.2603, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1365.5409, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1573.5469, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1318.4055, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(863.3854, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(888.4546, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(651.1584, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1009.8204, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1069.4189, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1274.2666, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1291.4220, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1657.5262, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(562.3496, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(906.2300, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(998.8359, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(708.4658, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1072.8126, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1193.1509, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(501.4961, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1182.1094, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(731.1270, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1126.7490, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(659.9086, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(511.1266, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(857.0984, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1451.4614, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(885.9369, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1502.3707, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(583.5482, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(809.2422, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(912.2359, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(880.1241, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(825.8856, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(913.5046, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(595.2850, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(774.6010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1621.5762, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1533.3009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1336.3074, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(542.9697, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(753.4576, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(833.3434, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(838.0024, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.15337423312883436, 'recall': 0.09881422924901186, 'f1': 0.1201923076923077, 'number': 253}, 'P': {'precision': 0.1473922902494331, 'recall': 0.18950437317784258, 'f1': 0.16581632653061226, 'number': 343}, 'overall_precision': 0.1490066225165563, 'overall_recall': 0.15100671140939598, 'overall_f1': 0.15, 'overall_accuracy': 0.5878970619171167}
			------------EPOCH 5---------------
Loss:  tensor(755.2142, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(835.9054, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(888.9512, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(385.6525, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(682.2693, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(453.7368, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(366.1403, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(381.8666, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(614.7339, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(378.2832, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(982.4539, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1289.5337, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(544.9710, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(744.2837, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1376.7391, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2170.4756, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(931.5364, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1124.4690, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1022.1736, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(655.1895, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(609.8411, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(497.3428, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(723.1724, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(838.8405, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(876.3447, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(981.5352, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1396.8438, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(422.5411, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(680.7278, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(801.8572, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(591.4827, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(860.6589, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1008.4694, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(386.6271, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1107.6201, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(554.6664, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(908.1715, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(426.9499, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(316.3365, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(587.5602, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1046.3499, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(591.5084, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1053.0667, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(399.6775, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(614.0258, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(715.5198, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(743.1831, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(634.0039, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(790.8044, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(590.1635, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(621.7914, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1483.5378, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2012.7997, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1115.3137, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(458.3921, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(577.1383, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(599.8256, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(579.8304, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.13240418118466898, 'recall': 0.15019762845849802, 'f1': 0.14074074074074075, 'number': 253}, 'P': {'precision': 0.1526479750778816, 'recall': 0.14285714285714285, 'f1': 0.14759036144578314, 'number': 343}, 'overall_precision': 0.14309210526315788, 'overall_recall': 0.14597315436241612, 'overall_f1': 0.14451827242524917, 'overall_accuracy': 0.5733819507748404}
			------------EPOCH 6---------------
Loss:  tensor(519.1801, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(680.5735, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(711.8682, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(335.2792, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(602.8842, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(345.4160, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(274.1541, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(275.6699, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(541.0381, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(278.2571, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1163.2251, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1637.7412, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(360.9760, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(785.8450, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1500.7074, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2687.0120, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(974.3895, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1258.3357, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(974.7774, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(738.9435, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(447.2439, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(368.1451, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(579.8080, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(668.8029, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1025.8999, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(811.3754, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1365.4229, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(497.0743, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(727.7084, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(877.8477, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(770.6777, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1043.1685, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1443.0857, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(397.1854, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1152.5701, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(554.8971, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1107.4138, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(431.0679, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(267.1674, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(590.4075, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(859.0575, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(617.5262, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1051.8190, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(363.9338, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(464.8500, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(525.7139, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(618.2511, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(514.5396, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(717.0657, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(455.6360, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(459.6324, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1202.2639, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1610.2297, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1396.1858, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(532.6912, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(582.6024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(798.0933, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(737.6256, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.06666666666666667, 'recall': 0.03162055335968379, 'f1': 0.04289544235924933, 'number': 253}, 'P': {'precision': 0.12538699690402477, 'recall': 0.23615160349854228, 'f1': 0.16380182002022245, 'number': 343}, 'overall_precision': 0.11618798955613577, 'overall_recall': 0.1493288590604027, 'overall_f1': 0.13069016152716595, 'overall_accuracy': 0.6174181333707314}
			------------EPOCH 7---------------
Loss:  tensor(850.5056, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(648.8246, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(905.4041, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(457.0438, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(575.4857, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(301.9741, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(272.4625, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(223.8640, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(484.2476, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(298.7286, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(812.2891, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1114.3568, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(353.8458, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(797.4619, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1239.9371, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2123.7876, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(807.6218, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(947.6584, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1027.6841, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(538.7833, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(483.1554, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(369.3552, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(534.0365, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(507.8308, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(646.6351, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(693.5905, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1080.9939, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(321.5965, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(568.6460, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(606.6037, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(374.9575, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(748.1503, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(654.6678, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(297.2835, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(892.7360, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(377.5568, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(858.8214, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(316.2862, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(172.8174, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(464.3747, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(900.3553, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(623.2086, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(938.5197, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(355.4795, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(472.4825, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(589.7886, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(713.8032, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(659.0584, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(663.9075, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(441.6086, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(540.1681, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1275.2258, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1173.9285, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(892.8843, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(316.0504, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(374.4344, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(446.1722, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(414.1104, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.1559633027522936, 'recall': 0.13438735177865613, 'f1': 0.14437367303609344, 'number': 253}, 'P': {'precision': 0.1634446397188049, 'recall': 0.27113702623906705, 'f1': 0.20394736842105263, 'number': 343}, 'overall_precision': 0.1613722998729352, 'overall_recall': 0.21308724832214765, 'overall_f1': 0.18365871294287778, 'overall_accuracy': 0.6456770212467569}
			------------EPOCH 8---------------
Loss:  tensor(388.5785, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(371.6468, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(431.3899, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(211.2286, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(427.4965, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(301.6980, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(175.0926, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(215.1480, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(367.6159, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(179.9531, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(725.0203, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1033.7812, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(172.5582, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(422.8930, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(950.5466, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1626.8445, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(659.2264, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(792.1943, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(713.6741, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(434.9614, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(352.9713, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(219.5550, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(389.2484, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(503.8152, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(425.3553, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(576.0973, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(997.1199, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(292.3171, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(342.9038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(454.2202, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(233.0201, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(492.8976, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(382.3948, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(179.5489, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(677.2983, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(252.3792, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(571.8058, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(224.7616, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(93.8351, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(254.3984, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(574.4674, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(461.4886, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(716.7266, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(234.7255, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(317.1182, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(389.2341, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(581.7200, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(385.0381, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(495.8677, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(354.4352, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(398.7362, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1069.3735, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1466.6726, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(803.0579, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(234.3267, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(317.4472, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(445.0405, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(418.0655, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.1576923076923077, 'recall': 0.16205533596837945, 'f1': 0.15984405458089668, 'number': 253}, 'P': {'precision': 0.23208191126279865, 'recall': 0.3965014577259475, 'f1': 0.29278794402583425, 'number': 343}, 'overall_precision': 0.20921985815602837, 'overall_recall': 0.29697986577181207, 'overall_f1': 0.24549237170596394, 'overall_accuracy': 0.635509431316177}
			------------EPOCH 9---------------
Loss:  tensor(408.7866, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(413.9615, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(359.4163, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(233.2703, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(487.4249, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(280.5375, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(184.6452, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(171.4042, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(421.2669, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(199.4594, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(667.5461, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(928.6940, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(163.9756, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(383.0995, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(905.9977, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1448.1628, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(470.1948, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(585.6420, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(511.0112, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(272.0330, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(230.7406, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(171.9936, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(244.4435, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(271.1947, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(297.2899, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(339.8022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(771.6600, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(227.9398, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(317.7780, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(409.3645, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(236.4553, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(516.1043, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(478.3270, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(178.4094, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(738.8354, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(303.1245, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(624.0780, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(262.3605, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(125.9757, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(357.0060, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(535.3535, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(480.9485, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(858.8125, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(236.1315, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(304.4717, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(321.5945, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(418.9859, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(378.4871, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(444.5398, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(268.7173, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(344.5276, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(683.4716, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(878.6716, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(649.3694, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(237.8453, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(229.4539, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(341.9893, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(314.1312, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.14022988505747128, 'recall': 0.24110671936758893, 'f1': 0.17732558139534885, 'number': 253}, 'P': {'precision': 0.1970954356846473, 'recall': 0.27696793002915454, 'f1': 0.23030303030303031, 'number': 343}, 'overall_precision': 0.17011995637949837, 'overall_recall': 0.26174496644295303, 'overall_f1': 0.20621282220753473, 'overall_accuracy': 0.6595610406002385}
			------------EPOCH 10---------------
Loss:  tensor(351.4441, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(315.5568, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(341.6646, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(177.2925, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(455.5140, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(225.9785, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(136.5144, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(151.9717, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(335.3575, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(198.3433, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(560.8244, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(779.8176, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(377.7012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(337.0822, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(890.8325, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1600.7871, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(467.0932, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(556.2954, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(488.8447, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(300.4693, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(347.9774, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(246.7478, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(512.5792, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(727.5297, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(660.2798, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(438.2945, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(888.2172, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(281.4008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(277.6926, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(357.2639, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(189.7295, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(397.2344, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(265.3905, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(98.4274, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(545.7336, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(135.3621, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(403.7455, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(184.6724, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.6818, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(170.7387, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(363.4936, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(320.9375, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(532.7821, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(142.4510, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(295.1443, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(341.3878, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(445.5340, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(418.7618, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(768.9003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(426.1188, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(570.7581, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(883.9818, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1132.3793, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(873.8417, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(324.8282, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(419.9388, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(345.1632, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(335.3196, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.09788867562380038, 'recall': 0.2015810276679842, 'f1': 0.1317829457364341, 'number': 253}, 'P': {'precision': 0.10740740740740741, 'recall': 0.08454810495626822, 'f1': 0.09461663947797716, 'number': 343}, 'overall_precision': 0.1011378002528445, 'overall_recall': 0.1342281879194631, 'overall_f1': 0.11535688536409518, 'overall_accuracy': 0.6079517565388122}
			------------EPOCH 11---------------
Loss:  tensor(230.3549, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(278.4896, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(346.1870, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(143.2413, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(351.4694, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(104.6194, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(99.2073, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(95.7083, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(162.0634, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(94.3024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(378.1300, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(595.3418, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(78.6038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(241.2769, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(538.0859, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(958.6930, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(347.6913, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(423.2659, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(354.9689, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(162.5176, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(190.3621, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(116.1623, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(188.4950, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(275.8802, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(355.3101, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(251.3134, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(752.8538, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(185.5294, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(322.8357, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(438.7166, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(236.0769, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(587.2978, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(420.1691, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(219.5291, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(828.0933, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(256.5247, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(549.8447, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(213.5791, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(94.6322, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(346.9991, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(572.8141, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(364.4543, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(546.0008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(165.7899, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(185.2200, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(199.3676, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(276.4901, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(292.8083, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(277.5573, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(180.1025, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(294.2123, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(674.6775, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(837.3919, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(598.7188, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(155.9684, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(161.5820, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(292.5516, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(236.1025, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.1226158038147139, 'recall': 0.3557312252964427, 'f1': 0.18237082066869303, 'number': 253}, 'P': {'precision': 0.102803738317757, 'recall': 0.03206997084548105, 'f1': 0.04888888888888889, 'number': 343}, 'overall_precision': 0.12009512485136742, 'overall_recall': 0.16946308724832215, 'overall_f1': 0.14057063326374392, 'overall_accuracy': 0.5038216113877008}
			------------EPOCH 12---------------
Loss:  tensor(466.8391, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(444.0045, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(756.4271, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(189.1744, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(464.7743, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(297.4791, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(131.0479, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(209.2416, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(255.9701, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(222.7391, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(683.8829, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1559.9773, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(100.2072, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(356.1068, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(831.9175, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1244.7437, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(410.7519, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(505.1396, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(402.0796, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(182.3970, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(184.5924, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(118.1890, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(191.0728, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(168.5192, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(239.5366, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(235.2256, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(582.4341, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(138.9866, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(243.1418, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(366.4906, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(216.4693, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(367.1856, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(421.3903, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(195.7790, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(617.8202, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(211.4189, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(435.5622, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(321.7289, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(471.6646, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(635.9041, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(851.1395, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(401.7000, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(815.7667, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(432.4306, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(509.1115, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(571.7982, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(663.6472, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(816.9659, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(509.3833, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(290.4862, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(406.7482, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1000.1973, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(919.0446, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(811.6984, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(168.5434, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(289.4022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(226.7951, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(231.5203, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.1614457831325301, 'recall': 0.2648221343873518, 'f1': 0.20059880239520955, 'number': 253}, 'P': {'precision': 0.1780821917808219, 'recall': 0.18950437317784258, 'f1': 0.18361581920903955, 'number': 343}, 'overall_precision': 0.16923076923076924, 'overall_recall': 0.2214765100671141, 'overall_f1': 0.19186046511627908, 'overall_accuracy': 0.6200827431456419}
			------------EPOCH 13---------------
Loss:  tensor(274.1414, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(233.4658, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(191.2197, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(82.4448, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(389.4780, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(212.9901, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(110.9179, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(137.1059, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(267.3483, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(167.9819, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(497.3153, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(820.7342, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(323.0287, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(478.1475, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(889.0287, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1383.3708, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(500.3239, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(670.6503, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(718.7548, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(459.6704, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(253.1168, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(194.7263, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(275.6099, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(435.7911, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(491.4666, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(268.7912, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(516.1987, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(175.7173, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(230.1274, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(332.7921, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(193.2894, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(310.3862, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(246.5057, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(65.5784, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(430.4473, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(120.7243, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(276.3188, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(134.1329, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.2740, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(164.9008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(355.0389, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(212.5805, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(462.6673, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(115.6013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(208.4595, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(304.2000, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(310.2502, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(258.4519, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(377.3848, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(180.8138, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(204.2565, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(576.0934, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(748.9640, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(715.8589, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(268.0835, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(246.5125, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(472.8347, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(458.3394, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.2, 'recall': 0.12648221343873517, 'f1': 0.1549636803874092, 'number': 253}, 'P': {'precision': 0.20234113712374582, 'recall': 0.35276967930029157, 'f1': 0.25717321997874604, 'number': 343}, 'overall_precision': 0.20184696569920843, 'overall_recall': 0.25671140939597314, 'overall_f1': 0.2259970457902511, 'overall_accuracy': 0.6575275226141224}
			------------EPOCH 14---------------
Loss:  tensor(333.3879, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(311.5420, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(303.3475, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(180.4121, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(291.3960, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(116.1284, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(80.1835, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(105.4715, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(167.0925, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(62.7089, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(319.8056, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(494.9827, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.7152, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(215.1703, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(607.7717, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(941.6458, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(288.0795, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(367.1341, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(237.7570, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(108.4684, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(118.2692, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(69.0767, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(123.2366, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(131.1255, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(152.7416, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(91.5122, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(377.6560, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(102.4629, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(156.5795, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(254.7637, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(97.5538, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(266.2305, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(198.3409, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(67.9526, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(369.9302, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(70.9562, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(263.6089, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(96.7489, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.9228, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(120.8465, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(312.1922, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(208.4368, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(306.1890, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(94.5932, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(134.4059, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(123.9690, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(203.6970, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(201.4420, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(196.0248, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(114.3479, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(170.1682, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(439.4677, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(576.0323, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(390.2709, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(89.9530, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(95.7161, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(120.4873, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(96.2920, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.13953488372093023, 'recall': 0.23715415019762845, 'f1': 0.17569546120058566, 'number': 253}, 'P': {'precision': 0.16990291262135923, 'recall': 0.20408163265306123, 'f1': 0.18543046357615894, 'number': 343}, 'overall_precision': 0.1543942992874109, 'overall_recall': 0.2181208053691275, 'overall_f1': 0.18080667593880387, 'overall_accuracy': 0.6441343524297034}
			------------EPOCH 15---------------
Loss:  tensor(151.9257, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(148.6076, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(132.2013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.8195, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(150.1403, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.4506, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.2924, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.7250, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(103.5676, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.0420, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(230.7277, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(417.1645, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.9584, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(123.7884, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(374.7265, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(686.4153, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(242.1773, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(302.0531, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(199.7509, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(81.4928, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(79.7678, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.3507, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(64.1692, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(94.3744, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(73.1620, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(62.0942, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(300.0313, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.9661, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(91.7635, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(185.0239, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(65.7862, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(187.6113, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(126.7884, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.4694, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(242.9670, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.1495, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(149.8136, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(70.2626, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.9109, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(70.7788, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(219.3771, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(115.1053, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(191.9540, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.3477, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(98.0154, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(109.3663, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(203.5162, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(165.3127, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(135.9879, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(69.1976, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(72.8257, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(320.5261, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(542.9852, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(362.2000, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(66.6332, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(72.8223, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(84.4106, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(68.4436, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.19927536231884058, 'recall': 0.21739130434782608, 'f1': 0.20793950850661627, 'number': 253}, 'P': {'precision': 0.2032854209445585, 'recall': 0.2886297376093295, 'f1': 0.23855421686746986, 'number': 343}, 'overall_precision': 0.2018348623853211, 'overall_recall': 0.25838926174496646, 'overall_f1': 0.22663723325974983, 'overall_accuracy': 0.659140312741042}
			------------EPOCH 16---------------
Loss:  tensor(106.8004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(109.4865, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(74.3528, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.0970, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(114.6079, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.1292, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.2519, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.5887, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(72.9144, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.6205, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(192.2000, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(347.5898, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.2445, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(84.7475, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(315.0944, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(606.5315, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(192.7664, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(251.9748, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(156.8533, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.2892, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.0012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.4108, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.3595, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(64.8929, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.3573, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.1239, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(249.7913, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.2109, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(59.9967, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(159.3458, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.2308, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(133.5830, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(122.3033, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.0483, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(200.4657, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.9426, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(116.5186, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.9532, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.6155, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.0692, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(201.4353, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(70.1940, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(137.9561, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.8079, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(65.9202, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(62.6476, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(144.4681, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(125.5020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(89.7780, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.9396, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.1249, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(252.0694, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(472.5051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(276.4776, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.1595, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.4657, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.6266, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.1504, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.1942675159235669, 'recall': 0.24110671936758893, 'f1': 0.21516754850088185, 'number': 253}, 'P': {'precision': 0.21017699115044247, 'recall': 0.27696793002915454, 'f1': 0.2389937106918239, 'number': 343}, 'overall_precision': 0.20365535248041775, 'overall_recall': 0.26174496644295303, 'overall_f1': 0.22907488986784144, 'overall_accuracy': 0.6547927915293458}
			------------EPOCH 17---------------
Loss:  tensor(83.3906, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(86.2053, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.6002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.5161, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(84.5106, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.7995, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.1925, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.7040, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.6217, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.8133, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(158.2259, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(309.8192, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.9535, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(71.5539, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(273.2065, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(570.8968, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(163.3035, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(197.1973, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(112.1945, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.7731, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.1115, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.1146, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.6222, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.6260, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.6175, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.0199, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(209.1120, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.7303, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.1079, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(135.1075, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.9081, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(109.1625, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(105.2252, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.4531, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(164.2715, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.8910, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(91.5174, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.7768, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.4241, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.2652, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(174.2482, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.6454, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(103.2934, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.8280, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.8946, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.9668, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(120.8596, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(100.4774, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(66.6970, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.4238, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.7942, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(220.0344, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(447.7282, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(233.3632, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.2140, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.9946, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.8868, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.9570, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.19218241042345277, 'recall': 0.233201581027668, 'f1': 0.21071428571428572, 'number': 253}, 'P': {'precision': 0.1961206896551724, 'recall': 0.2653061224489796, 'f1': 0.22552664188351917, 'number': 343}, 'overall_precision': 0.19455252918287938, 'overall_recall': 0.2516778523489933, 'overall_f1': 0.21945866861741042, 'overall_accuracy': 0.6585793422621135}
			------------EPOCH 18---------------
Loss:  tensor(63.8408, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(66.6007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.0159, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.7959, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(67.8852, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.5516, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.4218, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.7514, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.8204, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.4066, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(143.2451, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(283.8507, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.6381, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.0366, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(225.8023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(516.6053, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(145.6453, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(172.2057, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(92.4184, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.6366, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.0728, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.2941, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.3055, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.0382, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.9067, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.7670, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(189.5470, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.2115, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.4305, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(120.2381, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.7336, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(88.8701, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(87.4977, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.7112, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(146.5007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.2826, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(69.7312, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.6496, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.8618, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.8995, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(133.7910, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.4667, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(78.4239, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.2850, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.5943, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.5113, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(114.9208, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(88.2838, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.3253, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.3688, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.1969, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(185.0876, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(418.6556, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(203.2954, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.9649, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.5507, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.0187, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.1678, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.20634920634920634, 'recall': 0.25691699604743085, 'f1': 0.22887323943661972, 'number': 253}, 'P': {'precision': 0.20430107526881722, 'recall': 0.27696793002915454, 'f1': 0.23514851485148516, 'number': 343}, 'overall_precision': 0.20512820512820512, 'overall_recall': 0.2684563758389262, 'overall_f1': 0.23255813953488372, 'overall_accuracy': 0.6602622536988991}
			------------EPOCH 19---------------
Loss:  tensor(53.4979, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.8504, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.9225, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.6488, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.8151, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.6083, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.2108, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.1071, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.4779, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.8829, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(128.7330, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(259.2021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.3739, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.8016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(202.0135, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(475.4699, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(130.9345, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(150.2998, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(82.1097, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.9459, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.1536, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.4638, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.1396, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.4854, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.7838, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.7298, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(178.4303, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.5506, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.6886, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(110.1922, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.8798, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(66.1070, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(80.3418, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.6259, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(130.4275, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.8891, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.1221, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.2393, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.1620, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.0392, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(110.9119, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.8189, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.2408, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.4502, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.5090, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.8626, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(101.3023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(77.7087, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.2715, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.4373, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.3609, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(181.8804, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(397.2930, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(177.2523, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.8212, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.1620, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.3519, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.5876, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.20241691842900303, 'recall': 0.2648221343873518, 'f1': 0.22945205479452058, 'number': 253}, 'P': {'precision': 0.20776255707762556, 'recall': 0.2653061224489796, 'f1': 0.23303457106274011, 'number': 343}, 'overall_precision': 0.20546163849154747, 'overall_recall': 0.2651006711409396, 'overall_f1': 0.2315018315018315, 'overall_accuracy': 0.6660122011079167}
			------------EPOCH 20---------------
Loss:  tensor(39.1992, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.7163, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.8396, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.4736, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.6794, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.0575, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.6525, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.5140, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.2225, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.2905, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(115.1850, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(240.7987, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.9240, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.3079, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(181.1081, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(496.9145, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(183.4915, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(190.1505, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(93.6032, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.1130, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.3639, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.3093, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.8981, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.5757, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.7501, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.9370, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(165.6978, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.1941, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.5602, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(101.6254, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.6865, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.0779, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(73.6900, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.6309, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(121.7405, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.4863, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.4099, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.4049, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.3388, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.3381, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(125.4258, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.2970, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.7526, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.7940, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.7190, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.4144, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(101.9191, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(77.3256, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.7170, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.3173, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.2700, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(169.3095, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(401.2003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(172.6683, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.4819, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.1983, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.6707, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.7540, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.23247232472324722, 'recall': 0.2490118577075099, 'f1': 0.24045801526717558, 'number': 253}, 'P': {'precision': 0.19387755102040816, 'recall': 0.27696793002915454, 'f1': 0.22809123649459784, 'number': 343}, 'overall_precision': 0.2076215505913272, 'overall_recall': 0.2651006711409396, 'overall_f1': 0.2328666175386883, 'overall_accuracy': 0.6564757029661314}
			------------EPOCH 21---------------
Loss:  tensor(33.7598, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.6087, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.8230, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.1647, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.0133, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.6775, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.8714, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.0032, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.4540, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.1024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(111.9298, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(235.5087, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.9368, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.8194, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(170.5956, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(522.3895, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(224.0420, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(270.0275, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(151.7119, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.8685, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.0374, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.1151, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.4263, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.5124, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.7184, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.3349, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(160.5214, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.3111, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.8055, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(100.8810, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.2059, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.7536, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(87.9373, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.3091, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(130.9099, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.2226, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.1596, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.9446, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.3365, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.2475, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(149.3426, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.3964, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.3029, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.6587, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.4711, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.7414, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(117.0190, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(81.9626, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.5752, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.2723, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.9092, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(188.0946, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(447.0038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(166.9393, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.7484, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.6614, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.2952, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.9599, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.2147239263803681, 'recall': 0.2766798418972332, 'f1': 0.24179620034542312, 'number': 253}, 'P': {'precision': 0.2024793388429752, 'recall': 0.2857142857142857, 'f1': 0.2370012091898428, 'number': 343}, 'overall_precision': 0.2074074074074074, 'overall_recall': 0.28187919463087246, 'overall_f1': 0.23897581792318634, 'overall_accuracy': 0.6555641259378725}
			------------EPOCH 22---------------
Loss:  tensor(38.7645, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.7889, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.7766, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.8683, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.7467, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.8965, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.2496, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.9289, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.5351, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.3232, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(104.4226, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(226.9563, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.8664, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.6686, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(159.0472, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(547.3660, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(273.3131, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(232.8325, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(92.1597, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.8061, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.0405, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.0856, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.9648, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.7589, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.9355, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.5220, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(152.2425, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.7756, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.2370, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(101.3161, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.8338, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.4989, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(77.7452, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.8999, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(132.5376, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.4220, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.2623, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.8160, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.2528, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.8484, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(155.6146, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.7939, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.0629, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.4720, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.1617, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.1232, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(111.8997, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(85.6011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.8329, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.6456, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.4525, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(150.0977, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(419.8984, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(151.2195, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.0607, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.6562, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.6538, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.4593, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.21262458471760798, 'recall': 0.25296442687747034, 'f1': 0.23104693140794225, 'number': 253}, 'P': {'precision': 0.194, 'recall': 0.282798833819242, 'f1': 0.23013048635824437, 'number': 343}, 'overall_precision': 0.2009987515605493, 'overall_recall': 0.2701342281879195, 'overall_f1': 0.23049391553328563, 'overall_accuracy': 0.6486221162611318}
			------------EPOCH 23---------------
Loss:  tensor(27.5063, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.2660, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.2211, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.8937, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.4631, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.3349, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.0813, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.2972, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.5197, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.2187, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(115.0294, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(233.4520, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.0824, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.1800, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(163.0515, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(445.1089, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(200.1490, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(176.5005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(70.9881, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.5773, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.2896, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.0790, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.6098, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.1869, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.2815, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.4405, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(161.6653, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.2557, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.3229, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(98.9438, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.4704, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.2259, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(62.1120, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.5553, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(122.4576, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.0692, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.4150, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.3961, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.3569, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.5883, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(96.0734, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.1814, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.9214, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.0203, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.6919, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.0202, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(102.1601, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(75.3459, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.3041, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.6474, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.0564, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(133.9747, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(377.8950, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(134.4237, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.2702, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.2858, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.2389, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.3210, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.24489795918367346, 'recall': 0.2845849802371542, 'f1': 0.26325411334552107, 'number': 253}, 'P': {'precision': 0.21153846153846154, 'recall': 0.2886297376093295, 'f1': 0.24414303329223186, 'number': 343}, 'overall_precision': 0.22440944881889763, 'overall_recall': 0.28691275167785235, 'overall_f1': 0.25184094256259204, 'overall_accuracy': 0.6544421849800154}
			------------EPOCH 24---------------
Loss:  tensor(20.8287, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.7737, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.3199, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.8053, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.0482, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.8763, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.4313, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.3578, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.3949, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.9773, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(104.9385, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(213.6713, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.7164, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.3611, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(153.4234, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(398.4379, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(130.6374, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(126.7643, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.9662, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.9181, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.9224, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.5070, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.3633, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.8938, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.4415, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.3892, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(147.9002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.4443, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.5359, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(87.4358, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.0782, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.7242, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(59.8006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.4131, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(102.4625, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.1341, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.8387, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.3515, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.7306, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.5593, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(77.1267, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.6039, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.3787, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.9219, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.4933, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.4223, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(74.4726, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(61.5143, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.4312, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.5478, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.9988, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(110.3455, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(348.8236, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(118.7045, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.4108, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.8545, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.8816, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.9568, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.23939393939393938, 'recall': 0.31225296442687744, 'f1': 0.27101200686106347, 'number': 253}, 'P': {'precision': 0.2096069868995633, 'recall': 0.27988338192419826, 'f1': 0.2397003745318352, 'number': 343}, 'overall_precision': 0.22208121827411167, 'overall_recall': 0.2936241610738255, 'overall_f1': 0.25289017341040465, 'overall_accuracy': 0.654021457120819}
			------------EPOCH 25---------------
Loss:  tensor(16.0406, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.2660, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.2811, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.7116, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.3977, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.2286, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.0833, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.4395, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.5576, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.0854, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(85.6871, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(191.6412, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.9851, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.9885, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(142.8461, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(361.2432, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(112.2916, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(110.9485, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.3638, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.5167, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.0041, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.0411, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.4505, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.9918, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.6838, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.1257, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(127.9647, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.0031, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.1000, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(75.2691, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.0483, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.5152, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.3520, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.7624, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(93.7933, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.4000, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.2026, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.3842, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.8901, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.1384, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.8802, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.1637, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.0488, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.4328, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.5349, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.6464, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(85.8750, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(59.4471, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.6214, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.4861, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.1168, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(111.4642, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(340.6960, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(117.9339, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.6734, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.3773, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.2017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.9580, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.24912280701754386, 'recall': 0.28063241106719367, 'f1': 0.2639405204460966, 'number': 253}, 'P': {'precision': 0.19631901840490798, 'recall': 0.27988338192419826, 'f1': 0.23076923076923078, 'number': 343}, 'overall_precision': 0.2157622739018088, 'overall_recall': 0.2802013422818792, 'overall_f1': 0.24379562043795622, 'overall_accuracy': 0.6576677652338546}
			------------EPOCH 26---------------
Loss:  tensor(15.8832, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.3002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.0033, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.0983, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.7705, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.9951, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.5737, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.4772, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.3962, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.0052, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(85.6998, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(182.2928, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.1621, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.3147, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(120.1362, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(323.0980, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(96.7018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(113.9894, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(66.6150, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.4007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.5475, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.9382, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.8193, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.2781, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.2643, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.0283, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(117.9869, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.3813, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.6401, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(81.3336, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.4445, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.6613, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.9775, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.0225, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(101.1682, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.9109, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.5942, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.2321, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.4129, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.5020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(62.6575, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.4215, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.1439, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.1815, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.6000, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.3907, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(103.9166, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(65.3751, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.0593, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.6823, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.1955, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(104.4063, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(316.8672, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(106.5763, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.0279, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.4975, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.9226, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.2489, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.22866894197952217, 'recall': 0.2648221343873518, 'f1': 0.24542124542124544, 'number': 253}, 'P': {'precision': 0.17365269461077845, 'recall': 0.2536443148688047, 'f1': 0.20616113744075829, 'number': 343}, 'overall_precision': 0.19395465994962216, 'overall_recall': 0.25838926174496646, 'overall_f1': 0.22158273381294966, 'overall_accuracy': 0.6510062407965781}
			------------EPOCH 27---------------
Loss:  tensor(15.6741, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.9754, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.8698, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.7178, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.9637, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.0466, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.9185, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.4282, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.7472, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.4339, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(84.4473, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(179.3033, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.2848, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.3253, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(120.4395, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(288.3232, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(80.0341, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(88.8903, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.8911, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.1010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.7375, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.6466, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.9370, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.4894, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.4646, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.0594, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(142.8791, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.1595, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(92.1151, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.6899, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.7154, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.8870, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.1782, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(102.7169, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.6433, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.0837, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.7914, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.4495, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.0433, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.6808, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.9535, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.5072, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.2640, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.2331, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.0841, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(88.7108, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(71.9415, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.9566, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.7758, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.4638, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(100.1773, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(288.9625, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(98.3607, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.3240, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.7642, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.4196, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.6509, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.23548387096774193, 'recall': 0.2885375494071146, 'f1': 0.25932504440497334, 'number': 253}, 'P': {'precision': 0.21238938053097345, 'recall': 0.27988338192419826, 'f1': 0.24150943396226415, 'number': 343}, 'overall_precision': 0.22178477690288714, 'overall_recall': 0.2835570469798658, 'overall_f1': 0.24889543446244478, 'overall_accuracy': 0.6546525489096137}
			------------EPOCH 28---------------
Loss:  tensor(16.2373, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.6315, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.3427, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.4265, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.1955, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.9122, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.7993, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.6665, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.4402, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2946, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(69.4979, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(150.5214, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.5815, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.1210, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(111.6220, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(305.9620, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(87.1711, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(78.8628, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.8352, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.9097, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.4502, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.3515, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.7702, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.2050, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.1297, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.5989, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(147.5600, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.1749, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.8055, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(68.3115, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.8045, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.5956, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.7929, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.0859, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(92.4375, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.3759, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.0859, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.5460, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.8926, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.8016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.8429, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.2562, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.6912, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.2452, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.1381, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.0923, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(62.5299, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(59.8275, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.3411, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.0897, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.1091, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(102.2126, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(315.3726, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(98.9579, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.8050, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.5505, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.0560, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.5645, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.2692307692307692, 'recall': 0.30434782608695654, 'f1': 0.28571428571428575, 'number': 253}, 'P': {'precision': 0.2222222222222222, 'recall': 0.30903790087463556, 'f1': 0.25853658536585367, 'number': 343}, 'overall_precision': 0.2398427260812582, 'overall_recall': 0.3070469798657718, 'overall_f1': 0.2693156732891832, 'overall_accuracy': 0.6558446111773368}
			------------EPOCH 29---------------
Loss:  tensor(8.7459, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.8732, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.2133, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.2721, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.9818, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.4702, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.4135, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.0928, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.8006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.4833, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(83.7346, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(185.9483, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.2482, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.4285, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(120.2964, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(336.5134, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(90.3139, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(74.7350, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.1834, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.9792, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.8422, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.4365, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.5000, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.7851, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.3073, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.6664, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(104.4071, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.4174, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.8654, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(77.1690, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.3092, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.1490, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(75.6666, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.9235, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(95.6533, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.7210, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.1255, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.5644, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.3585, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.0121, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(90.3217, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.6562, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.1025, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.3205, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.9557, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.6291, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(110.5870, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.1100, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.2817, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.4535, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.4249, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(134.8741, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(388.9286, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(118.9309, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.5705, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.2541, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.6286, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.8754, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.23809523809523808, 'recall': 0.31620553359683795, 'f1': 0.27164685908319186, 'number': 253}, 'P': {'precision': 0.2234513274336283, 'recall': 0.2944606413994169, 'f1': 0.2540880503144654, 'number': 343}, 'overall_precision': 0.2296954314720812, 'overall_recall': 0.3036912751677852, 'overall_f1': 0.26156069364161844, 'overall_accuracy': 0.6544421849800154}
			------------EPOCH 30---------------
Loss:  tensor(9.3052, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.5083, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.7761, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.4363, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.5326, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.3024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.1852, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.4265, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.8141, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.7786, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(82.6507, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(179.1001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.2888, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.7898, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(98.9377, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(264.7813, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(88.9096, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(81.1570, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.8293, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.9326, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.6557, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.9953, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.4860, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.9623, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.0110, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.1955, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(132.9702, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.3568, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.9699, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(79.9797, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.0254, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.3309, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(75.0072, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.7629, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(91.0306, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.0513, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.5954, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.8085, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.4790, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.9405, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(111.9929, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(67.5445, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.1029, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.4833, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.5074, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.0922, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(79.1454, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(62.8910, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.4849, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.9232, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.9312, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(114.9079, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(455.1025, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(96.2231, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.3821, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.0517, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.3766, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.4433, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.24067796610169492, 'recall': 0.28063241106719367, 'f1': 0.25912408759124084, 'number': 253}, 'P': {'precision': 0.1894093686354379, 'recall': 0.27113702623906705, 'f1': 0.22302158273381295, 'number': 343}, 'overall_precision': 0.20865139949109415, 'overall_recall': 0.2751677852348993, 'overall_f1': 0.23733719247467439, 'overall_accuracy': 0.6518476965149709}


		-------------RUN 2-----------
			------------EPOCH 1---------------
Loss:  tensor(2978.0898, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2445.7161, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2361.2581, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3368.8086, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4481.3540, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3322.5593, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3379.0725, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2211.6313, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1205.0731, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1505.7726, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1584.4282, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1521.6611, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2417.5215, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2565.3804, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1040.3456, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1522.2063, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2324.6621, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2422.5073, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1765.6909, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1259.4469, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2080.7244, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1820.1111, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1645.5911, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1748.6699, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1332.9189, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1935.0243, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1572.2349, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1169.4805, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1965.9641, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2009.7439, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2506.4009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1420.2258, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1293.5951, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2274.8262, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2225.5630, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1636.3503, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(965.3632, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(807.9401, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2905.7056, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2271.3779, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3137.1292, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1271.3226, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1949.6702, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1403.2914, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2725.2983, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3019.5701, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1671.8672, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2663.8284, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1131.9333, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(971.4056, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2005.6416, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(995.0645, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(764.3090, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2420.1675, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2614.4778, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2413.7544, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1765.7866, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.044563279857397504, 'recall': 0.12562814070351758, 'f1': 0.06578947368421052, 'number': 199}, 'P': {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'number': 297}, 'overall_precision': 0.04448398576512456, 'overall_recall': 0.05040322580645161, 'overall_f1': 0.04725897920604914, 'overall_accuracy': 0.35496970907396314}
			------------EPOCH 2---------------
Loss:  tensor(2264.9404, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1630.1702, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1712.5958, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2355.5779, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3502.6841, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2556.9353, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2573.2637, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1741.7041, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(966.9432, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1265.6152, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1288.6824, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1233.9988, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2011.0740, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2209.2886, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(884.7478, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1216.1409, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1753.7723, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2014.8113, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1516.8110, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1085.4724, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1941.0166, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1550.4622, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1467.3267, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1577.0696, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1020.8900, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1417.9487, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1143.8682, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1060.8018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1494.3145, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1512.8318, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2038.0549, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1196.3943, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1026.2859, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1840.2462, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1849.4019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1269.0597, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(763.0436, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(676.2615, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2471.3892, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1991.2485, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2576.8208, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1023.3027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1429.2119, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1201.4377, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2185.8564, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2682.9902, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1397.3514, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2304.4331, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(934.4976, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(791.9606, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1619.5959, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(854.5501, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(616.3319, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1995.0034, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2179.4309, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1916.4503, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1461.2323, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.08635578583765112, 'recall': 0.25125628140703515, 'f1': 0.12853470437017994, 'number': 199}, 'P': {'precision': 0.030303030303030304, 'recall': 0.003367003367003367, 'f1': 0.006060606060606061, 'number': 297}, 'overall_precision': 0.08333333333333333, 'overall_recall': 0.1028225806451613, 'overall_f1': 0.09205776173285198, 'overall_accuracy': 0.38486119432794086}
			------------EPOCH 3---------------
Loss:  tensor(1893.0292, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1247.1450, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1467.6370, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1920.7151, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2867.8467, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2280.2393, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2242.6653, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1461.8884, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(758.2797, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1014.7502, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1049.1757, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1023.6678, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1672.2916, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1951.1782, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(729.6532, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(911.5136, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1369.7129, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1625.3516, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1117.3909, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(864.1606, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1317.0459, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1305.9797, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1255.3240, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1330.1462, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(838.7628, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1174.2894, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(931.5077, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(841.6232, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1165.9585, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1173.4532, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1808.8103, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1036.5522, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(825.6169, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1496.1880, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1537.6974, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1018.3418, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(629.8255, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(523.6562, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1947.6965, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1670.3743, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1962.5222, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(764.9180, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1012.2919, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(975.8275, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1799.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2427.1230, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1087.4788, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1856.9639, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(815.0072, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(643.8347, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1308.7058, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(689.8201, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(479.3287, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1693.1167, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1836.9712, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1696.2040, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1351.8279, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.09983361064891846, 'recall': 0.3015075376884422, 'f1': 0.14999999999999997, 'number': 199}, 'P': {'precision': 0.05263157894736842, 'recall': 0.010101010101010102, 'f1': 0.016949152542372885, 'number': 297}, 'overall_precision': 0.09574468085106383, 'overall_recall': 0.12701612903225806, 'overall_f1': 0.10918544194107452, 'overall_accuracy': 0.43499101258238465}
			------------EPOCH 4---------------
Loss:  tensor(1702.1455, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(965.8627, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1205.9974, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1680.9375, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2395.0957, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1951.8560, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1760.4468, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1337.7175, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(550.5334, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(796.8475, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(874.3624, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(859.8695, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1420.6941, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1718.6440, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(658.9455, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(665.4728, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(985.2639, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1167.3367, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(742.8942, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(648.6606, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(826.1069, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(978.2113, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1042.3085, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1084.6321, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(654.4177, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(890.2626, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(742.7014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(683.1702, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(869.1603, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(921.8461, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1493.6091, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(880.0056, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(676.1292, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1214.9729, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1288.3098, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(784.2304, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(480.3559, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(429.5177, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1247.0056, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1429.6766, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1526.6086, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(493.1482, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(686.9724, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(715.8015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1534.0869, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1864.6731, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(868.5900, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1581.2106, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(856.3321, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(540.9471, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1100.3940, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(547.6434, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(419.8665, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1548.1907, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1724.6140, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1492.0625, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1154.0188, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.14625228519195613, 'recall': 0.4020100502512563, 'f1': 0.21447721179624663, 'number': 199}, 'P': {'precision': 0.06666666666666667, 'recall': 0.010101010101010102, 'f1': 0.01754385964912281, 'number': 297}, 'overall_precision': 0.14020270270270271, 'overall_recall': 0.16733870967741934, 'overall_f1': 0.15257352941176472, 'overall_accuracy': 0.42547100725650755}
			------------EPOCH 5---------------
Loss:  tensor(1499.0559, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(737.5734, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(963.9343, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1395.3538, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2442.0195, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1636.4436, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1476.1418, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1270.4711, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(462.6387, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(715.2221, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(735.2482, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(777.9041, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1021.7880, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1377.6941, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(543.7627, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(529.3425, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(770.0149, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(849.2902, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(532.4714, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(581.5376, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(748.4639, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1013.8785, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(858.4370, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(977.5825, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(595.1312, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(732.9550, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(643.8656, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(617.2480, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(798.6398, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(905.5475, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1143.0422, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(783.4067, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(754.3882, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1277.2788, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1235.8535, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(627.8185, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(443.9427, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(336.6836, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1001.6080, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1393.3088, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1256.1560, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(456.9257, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(624.8895, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(707.2085, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1161.7336, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1577.2704, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(917.0756, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1515.0555, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(720.5914, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(502.8132, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1015.7951, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(426.1617, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(299.3602, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1196.3528, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1309.4224, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1249.0752, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(935.2709, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.1931216931216931, 'recall': 0.36683417085427134, 'f1': 0.2530329289428076, 'number': 199}, 'P': {'precision': 0.12168141592920353, 'recall': 0.18518518518518517, 'f1': 0.14686248331108145, 'number': 297}, 'overall_precision': 0.15421686746987953, 'overall_recall': 0.25806451612903225, 'overall_f1': 0.19306184012066369, 'overall_accuracy': 0.6377072099061314}
			------------EPOCH 6---------------
Loss:  tensor(1386.6880, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1205.4058, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(759.8374, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1571.3936, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2243.5601, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1676.1482, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1618.5768, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1132.5104, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(381.1135, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(555.0876, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(673.5778, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(672.7339, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1363.7938, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1325.7810, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(654.7986, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(734.1226, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1140.4685, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1156.1113, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(382.6873, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(354.3205, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(477.6765, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(934.4361, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(987.9500, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(901.4201, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(570.0463, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(802.2444, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(688.1381, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(430.7104, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(741.9911, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(890.5382, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1092.2095, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(618.9796, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(534.4802, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(963.2329, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(928.4501, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(557.6679, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(428.6500, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(311.1541, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(912.4562, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1091.5405, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1227.0186, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(462.4835, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(638.7201, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(515.4960, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1113.6826, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1454.4242, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(730.1554, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1177.5209, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(603.5450, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(383.0972, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(762.9141, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(267.4686, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(247.6494, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1038.5743, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1142.4890, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1062.1204, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(685.1371, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.1561938958707361, 'recall': 0.4371859296482412, 'f1': 0.2301587301587302, 'number': 199}, 'P': {'precision': 0.07168458781362007, 'recall': 0.06734006734006734, 'f1': 0.06944444444444445, 'number': 297}, 'overall_precision': 0.12799043062200957, 'overall_recall': 0.2157258064516129, 'overall_f1': 0.16066066066066068, 'overall_accuracy': 0.5747952865987618}
			------------EPOCH 7---------------
Loss:  tensor(899.2772, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(404.8352, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(494.1049, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(916.6742, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1617.9048, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(997.2158, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(798.1134, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(906.7040, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(306.9471, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(394.4521, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(473.3104, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(496.0492, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(584.4838, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(775.6526, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(361.0718, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(329.6028, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(505.2957, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(592.2692, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(208.4406, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(225.7098, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(243.2734, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(512.4684, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(652.4818, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(605.7303, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(314.8386, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(361.4806, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(406.9835, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(263.6505, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(490.3160, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(519.1083, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(719.3478, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(463.7574, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(439.4624, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(712.3704, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(769.7140, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(396.5706, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(190.6121, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(230.1649, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(500.2607, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(661.7163, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(669.0765, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(260.9725, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(303.2500, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(284.6741, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(848.9197, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(963.3815, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(461.8528, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(745.2638, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(453.2708, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(274.9125, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(568.7145, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(183.9162, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(162.3094, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(829.1558, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(906.0302, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(832.0652, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(478.3038, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.19895287958115182, 'recall': 0.38190954773869346, 'f1': 0.26161790017211706, 'number': 199}, 'P': {'precision': 0.0891566265060241, 'recall': 0.12457912457912458, 'f1': 0.10393258426966291, 'number': 297}, 'overall_precision': 0.14178168130489335, 'overall_recall': 0.22782258064516128, 'overall_f1': 0.17478731631863884, 'overall_accuracy': 0.6519539311630385}
			------------EPOCH 8---------------
Loss:  tensor(655.0334, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(248.1858, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(357.5933, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(750.4265, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1274.3252, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(814.7157, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(626.4679, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(789.2866, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(245.6030, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(270.7904, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(354.7767, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(345.7051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(333.3266, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(587.7516, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(289.4737, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(252.2298, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(409.5297, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(565.5065, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(174.4170, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(135.0464, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(227.9792, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(498.7704, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(615.1100, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(469.2605, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(231.0910, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(297.3068, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(361.2179, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(178.7762, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(475.8291, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(495.3515, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(681.9446, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(437.9415, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(371.7832, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(648.2070, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(589.0551, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(312.7126, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(145.3161, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(187.9798, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(387.0085, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(431.6789, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(430.6884, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(179.9189, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(171.3810, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(202.1817, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(675.5264, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(668.8063, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(244.2518, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(535.5323, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(397.7567, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(270.8918, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(472.8309, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(158.8976, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(171.5118, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(888.8190, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(844.4520, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(841.7516, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(533.7954, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.2, 'recall': 0.22110552763819097, 'f1': 0.2100238663484487, 'number': 199}, 'P': {'precision': 0.136986301369863, 'recall': 0.2356902356902357, 'f1': 0.17326732673267323, 'number': 297}, 'overall_precision': 0.15595075239398085, 'overall_recall': 0.22983870967741934, 'overall_f1': 0.18581907090464544, 'overall_accuracy': 0.6615405099527328}
			------------EPOCH 9---------------
Loss:  tensor(746.8041, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(301.0946, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(448.8386, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(672.2676, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1206.6548, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(786.3622, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(586.7857, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(847.2352, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(228.5402, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(260.3874, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(389.9778, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(324.5809, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(204.8159, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(453.3326, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(201.6063, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(133.8849, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(302.6276, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(429.7309, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(137.3757, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(100.8946, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(192.1091, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(435.6122, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(484.5583, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(331.0888, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(233.1785, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(291.3701, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(343.8441, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(230.6086, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(710.3079, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(619.7852, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(896.1295, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(584.3334, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(388.8778, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(625.1151, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(712.3246, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(438.0979, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(227.4717, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(202.0550, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(427.1070, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(747.9324, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(550.0527, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(196.4432, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(181.7909, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(282.6335, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(654.3647, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(675.2980, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(191.7598, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(545.1063, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(278.5573, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(183.9484, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(407.8405, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(114.1234, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(143.4305, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(645.5051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(650.1672, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(706.6967, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(491.9691, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.2761904761904762, 'recall': 0.1457286432160804, 'f1': 0.19078947368421056, 'number': 199}, 'P': {'precision': 0.14683153013910355, 'recall': 0.31986531986531985, 'f1': 0.20127118644067796, 'number': 297}, 'overall_precision': 0.16489361702127658, 'overall_recall': 0.25, 'overall_f1': 0.1987179487179487, 'overall_accuracy': 0.6388389587910259}
			------------EPOCH 10---------------
Loss:  tensor(778.1750, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(483.4669, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(565.2507, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(844.1592, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1490.0747, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1116.2654, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1059.3535, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1146.2361, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(449.6146, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(438.9731, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(787.8425, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(725.8258, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(978.2909, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(903.6371, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(397.8745, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(372.3918, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(406.8330, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(389.8706, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(170.9754, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(239.7854, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(173.4474, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(425.0823, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(555.9586, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(280.3206, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(227.9088, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(289.6182, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(351.9203, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(136.7205, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(401.7403, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(385.0702, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(481.4024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(333.2963, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(405.0349, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(646.2676, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(628.4178, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(304.6556, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(357.5839, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(265.5229, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(671.5858, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(831.2701, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1026.8861, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(334.1973, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(547.7820, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(577.3962, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1350.3567, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1723.8055, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(756.4124, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1016.2667, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(350.6447, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(405.9052, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(958.1576, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(306.0164, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(207.0499, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(995.9573, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(994.6909, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1211.5985, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(448.2654, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.1390593047034765, 'recall': 0.3417085427135678, 'f1': 0.19767441860465115, 'number': 199}, 'P': {'precision': 0.05569007263922518, 'recall': 0.07744107744107744, 'f1': 0.0647887323943662, 'number': 297}, 'overall_precision': 0.1008869179600887, 'overall_recall': 0.18346774193548387, 'overall_f1': 0.1301859799713877, 'overall_accuracy': 0.6221955928366953}
			------------EPOCH 11---------------
Loss:  tensor(530.2953, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(425.1491, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(351.4117, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(773.4117, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1262.9543, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(687.6467, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(570.6086, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(624.1971, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(271.9941, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(338.6047, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(670.9443, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(566.2861, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1378.0459, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1603.1965, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(758.6752, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(593.7672, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(825.7252, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(755.8578, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(373.9265, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(492.8646, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(371.4875, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(560.6472, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(749.6821, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(490.9293, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(311.3071, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(318.6029, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(399.9629, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(261.3598, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(478.2754, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(470.3710, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(664.4498, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(345.0213, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(347.6509, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(626.0896, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(490.3170, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(293.3622, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(180.9541, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(236.3079, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(372.9203, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(396.0895, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(524.2123, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(184.2901, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(199.5866, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(223.2812, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(668.2157, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(669.4241, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(277.9712, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(561.2035, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(293.5191, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(265.7981, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(602.3007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(186.9957, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(126.2602, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(844.3481, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(831.4733, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(958.9075, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(540.6021, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.17055655296229802, 'recall': 0.47738693467336685, 'f1': 0.2513227513227513, 'number': 199}, 'P': {'precision': 0.2708333333333333, 'recall': 0.13131313131313133, 'f1': 0.17687074829931973, 'number': 297}, 'overall_precision': 0.19115549215406563, 'overall_recall': 0.2701612903225806, 'overall_f1': 0.22389306599832917, 'overall_accuracy': 0.5075560881432661}
			------------EPOCH 12---------------
Loss:  tensor(759.3746, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(299.9256, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(470.2291, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(714.7503, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1339.9318, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(830.0264, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(615.8430, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(601.7174, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(200.2382, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(247.3420, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(303.5789, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(272.9100, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(260.8073, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(452.7949, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(187.4043, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(145.7728, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(295.1314, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(294.3228, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(159.3983, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(96.7181, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(145.2496, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(292.4672, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(315.3703, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(246.0735, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(163.7253, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(153.2606, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(212.8031, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(81.2589, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(250.7974, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(302.0256, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(374.3685, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(198.4545, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(327.4562, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(589.7794, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(465.6636, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(200.4466, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(116.3248, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(171.0446, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(326.3826, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(275.0316, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(367.5509, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(114.6226, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(98.9707, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(114.6595, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(605.8113, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(445.8364, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(109.9474, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(434.3710, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(156.7578, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(129.0350, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(254.5153, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(64.5832, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(80.8558, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(461.4066, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(443.6582, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(536.1080, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(225.7612, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.1434878587196468, 'recall': 0.32663316582914576, 'f1': 0.19938650306748468, 'number': 199}, 'P': {'precision': 0.043227665706051875, 'recall': 0.050505050505050504, 'f1': 0.04658385093167702, 'number': 297}, 'overall_precision': 0.1, 'overall_recall': 0.16129032258064516, 'overall_f1': 0.12345679012345678, 'overall_accuracy': 0.6138739098595299}
			------------EPOCH 13---------------
Loss:  tensor(373.9105, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(143.6467, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(219.2997, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(386.5093, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(805.8442, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(503.9578, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(330.4380, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(421.7192, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(122.9952, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(159.5566, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(192.3293, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(214.8897, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(108.0106, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(318.4361, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(136.8604, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(78.5870, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(168.6219, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(203.2924, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(103.1776, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.2155, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(62.4301, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(198.4626, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(252.7719, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(145.4116, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(114.7695, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(108.6592, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(147.6428, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.3366, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(222.7083, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(256.0568, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(245.3957, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(111.1036, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(292.2537, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(481.3121, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(326.1253, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(149.0525, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(114.6913, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(125.9692, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(267.7133, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(164.3108, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(226.9432, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(97.9325, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(93.9130, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(124.0083, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(472.4714, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(484.3567, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(111.6607, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(344.1491, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(139.8825, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(122.4869, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(203.6457, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.6691, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.0492, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(331.1674, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(356.3609, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(432.4742, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(154.7879, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.16009852216748768, 'recall': 0.32663316582914576, 'f1': 0.21487603305785125, 'number': 199}, 'P': {'precision': 0.049479166666666664, 'recall': 0.06397306397306397, 'f1': 0.05580029368575624, 'number': 297}, 'overall_precision': 0.10632911392405063, 'overall_recall': 0.1693548387096774, 'overall_f1': 0.13063763608087092, 'overall_accuracy': 0.6284534984355236}
			------------EPOCH 14---------------
Loss:  tensor(287.4343, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(91.8213, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(146.8563, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(288.7813, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(711.6206, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(422.1572, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(291.6292, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(348.7290, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(133.7437, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(129.4214, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(187.3255, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(144.1055, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(119.7834, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(306.8918, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(116.3159, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(92.2544, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(142.6187, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(179.2135, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(96.3813, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.1263, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(71.7419, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(260.8296, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(209.2584, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(171.6250, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(88.3338, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(77.7127, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(104.5639, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.0868, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(104.6712, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(128.4643, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(189.1857, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(72.0426, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(173.6091, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(289.6484, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(225.2508, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(72.0501, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(77.0109, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(104.0532, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(135.3116, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(67.0077, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(184.2903, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(70.4833, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.7220, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(97.8527, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(457.8373, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(357.2161, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(124.2643, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(312.6822, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(177.1046, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(139.4056, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(221.6666, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.3199, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.8856, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(411.2474, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(443.8420, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(579.3477, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(225.7312, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.16933638443935928, 'recall': 0.37185929648241206, 'f1': 0.2327044025157233, 'number': 199}, 'P': {'precision': 0.19827586206896552, 'recall': 0.23232323232323232, 'f1': 0.21395348837209302, 'number': 297}, 'overall_precision': 0.1821656050955414, 'overall_recall': 0.28830645161290325, 'overall_f1': 0.2232630757220921, 'overall_accuracy': 0.6367086079488716}
			------------EPOCH 15---------------
Loss:  tensor(282.2434, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(129.7079, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(146.3743, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(236.9814, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(601.4439, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(375.9819, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(229.7317, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(302.9092, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.3665, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(79.2126, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(119.3027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(93.0025, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(71.5599, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(224.7151, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(86.6383, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.5397, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(84.4155, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(161.9586, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(85.3193, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.3349, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(93.4914, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(213.9498, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(213.3736, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(169.9733, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(101.9683, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(124.7084, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(193.4670, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(96.6877, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(204.0281, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(217.4416, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(331.4376, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(184.5939, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(184.4752, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(382.6238, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(274.3799, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(107.2009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(71.1785, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(108.8472, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(129.2626, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(117.9968, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(198.8333, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.9148, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.9998, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.0810, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(371.8174, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(276.7428, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.8431, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(223.6024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(83.7817, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(90.7390, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(108.1739, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.1833, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.3667, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(262.6188, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(315.1711, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(399.0407, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(186.9550, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.18112244897959184, 'recall': 0.35678391959798994, 'f1': 0.24027072758037227, 'number': 199}, 'P': {'precision': 0.2641025641025641, 'recall': 0.3468013468013468, 'f1': 0.29985443959243085, 'number': 297}, 'overall_precision': 0.22250639386189258, 'overall_recall': 0.35080645161290325, 'overall_f1': 0.2723004694835681, 'overall_accuracy': 0.633912522468544}
			------------EPOCH 16---------------
Loss:  tensor(232.1642, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(140.2251, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(219.7322, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(235.2476, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(745.6846, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(565.2562, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(408.0156, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(412.4025, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(101.2862, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(122.5615, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(166.4200, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(211.9591, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(114.4170, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(357.9612, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(160.2697, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.9245, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(158.1322, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(152.0714, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(85.5073, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.1014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.6608, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(163.0469, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(223.0680, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(98.2480, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(86.2166, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(72.3448, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(84.0621, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.8782, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(80.1773, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(94.8864, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(150.7463, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(72.9832, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(132.2154, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(241.4296, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(179.7708, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(61.3221, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.8022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(103.5291, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(86.5927, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(115.9430, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(309.4351, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.8002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.8714, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(69.4278, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(432.1991, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(350.9284, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(86.0519, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(354.4879, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(100.6274, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(100.2190, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(206.2500, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(75.6933, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.9108, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(300.0155, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(358.9890, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(465.7241, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(102.7772, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.1650943396226415, 'recall': 0.35175879396984927, 'f1': 0.2247191011235955, 'number': 199}, 'P': {'precision': 0.1, 'recall': 0.11784511784511785, 'f1': 0.1081916537867079, 'number': 297}, 'overall_precision': 0.13565891472868216, 'overall_recall': 0.21169354838709678, 'overall_f1': 0.1653543307086614, 'overall_accuracy': 0.6295186738566008}
			------------EPOCH 17---------------
Loss:  tensor(196.3759, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(76.1272, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(110.3636, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(250.0448, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(618.4252, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(384.5087, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(299.7578, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(239.1020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.0037, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.0794, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(94.4869, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(73.3556, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.8009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(196.8397, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.8988, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.9412, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(77.1718, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(126.6993, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.3594, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.1178, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.5584, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(179.9578, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(177.7238, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(78.2299, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(79.6859, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(91.7702, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(99.2852, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.7631, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(110.8228, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(135.3596, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(175.4980, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.2868, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(219.6797, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(405.1045, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(244.0322, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(80.8708, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.3967, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.2934, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.5914, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(59.5093, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(224.0166, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.4970, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.9389, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.1402, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(408.5273, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(258.5953, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(65.7316, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(302.8911, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.8955, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(66.0403, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(79.1193, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.2135, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.0260, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(200.6004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(239.4805, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(290.8653, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(76.2506, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.17757009345794392, 'recall': 0.38190954773869346, 'f1': 0.24242424242424238, 'number': 199}, 'P': {'precision': 0.08947368421052632, 'recall': 0.11447811447811448, 'f1': 0.10044313146233383, 'number': 297}, 'overall_precision': 0.13613861386138615, 'overall_recall': 0.2217741935483871, 'overall_f1': 0.1687116564417178, 'overall_accuracy': 0.6423007789095266}
			------------EPOCH 18---------------
Loss:  tensor(178.9158, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.4174, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(79.2174, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(184.7817, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(532.7247, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(313.8684, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(216.2419, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(209.1002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.0059, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.4541, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(84.9909, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.1122, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.9764, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(149.5107, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.8879, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.8594, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.3282, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(113.1793, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.8162, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.3959, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.7484, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(135.5789, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(149.9591, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.4712, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.7688, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.0120, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.6544, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.3833, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(62.8078, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(79.4469, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(127.6916, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.1187, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(115.1635, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(214.4057, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(174.2556, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.2338, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.1220, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.2593, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.9745, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.9469, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(165.3639, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.0585, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.9785, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.4840, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(319.6024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(155.0537, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.9505, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(224.9397, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.1776, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.4824, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.2129, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.6207, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.8751, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(159.8002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(203.2675, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(252.4237, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.4518, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.1958762886597938, 'recall': 0.38190954773869346, 'f1': 0.25894378194207834, 'number': 199}, 'P': {'precision': 0.19174041297935104, 'recall': 0.21885521885521886, 'f1': 0.20440251572327042, 'number': 297}, 'overall_precision': 0.1939477303988996, 'overall_recall': 0.2842741935483871, 'overall_f1': 0.2305805396565822, 'overall_accuracy': 0.6520205046268558}
			------------EPOCH 19---------------
Loss:  tensor(135.7047, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.0450, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(70.6024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(142.0401, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(470.7379, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(284.0232, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(175.5394, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(174.7410, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.6587, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.2322, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(68.4193, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.7857, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.1595, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(123.5401, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.7044, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.8348, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.0383, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(86.4056, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.8485, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.1898, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.3802, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(93.3067, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(93.6766, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.1717, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.1047, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.8903, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.0372, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.0046, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.8615, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.8367, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(72.1639, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.7038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(85.4998, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(156.9080, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(118.0009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.8418, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.8669, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.7271, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.1766, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.0533, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(154.0656, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.9481, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.7089, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.0520, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(293.5131, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(144.1295, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.8818, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(201.7596, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.9072, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.3145, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.4539, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.4330, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.2867, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(146.5833, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(181.5730, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(218.1201, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.9962, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.18652849740932642, 'recall': 0.36180904522613067, 'f1': 0.24615384615384614, 'number': 199}, 'P': {'precision': 0.13903743315508021, 'recall': 0.1750841750841751, 'f1': 0.1549925484351714, 'number': 297}, 'overall_precision': 0.1631578947368421, 'overall_recall': 0.25, 'overall_f1': 0.19745222929936307, 'overall_accuracy': 0.6602090406763864}
			------------EPOCH 20---------------
Loss:  tensor(130.0052, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.5935, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(61.6414, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(134.9763, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(417.9735, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(235.7008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(146.7722, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(144.3314, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.5037, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.5496, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.0349, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.1794, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.3247, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(95.8838, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.5325, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.9460, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.0947, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(64.8103, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.7226, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.5085, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.8559, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(68.0564, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(83.4165, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.2942, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.8382, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.7515, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.8388, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.6324, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.6683, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.8309, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.5741, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.1964, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(61.7569, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(114.9780, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(75.6244, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.1896, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.0222, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.2890, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.0718, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.8151, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(137.0670, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.9515, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.9959, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.3167, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(265.2344, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(128.3173, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.7251, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(180.6357, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.8475, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.3375, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.5750, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.4673, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.2815, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(127.8209, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(162.3763, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(191.2318, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.0528, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.18717948717948718, 'recall': 0.36683417085427134, 'f1': 0.2478777589134126, 'number': 199}, 'P': {'precision': 0.16193181818181818, 'recall': 0.1919191919191919, 'f1': 0.17565485362095534, 'number': 297}, 'overall_precision': 0.1752021563342318, 'overall_recall': 0.2620967741935484, 'overall_f1': 0.210016155088853, 'overall_accuracy': 0.656414353238799}
			------------EPOCH 21---------------
Loss:  tensor(108.4197, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.8555, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.2469, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(122.2628, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(402.7267, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(216.5499, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(137.1518, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(125.0008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.6967, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.8560, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.5006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.8322, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.0875, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(82.7532, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.7947, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.8506, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.4752, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.4605, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.1968, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.9739, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.5510, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.6377, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(77.9816, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.8323, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.6418, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.4989, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.4894, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.9833, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.1727, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.4031, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.7820, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.8159, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.2402, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(99.1768, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(64.4283, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.9413, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.5201, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.0271, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.4054, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.2469, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(142.7700, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.0350, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.6732, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.1324, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(263.4024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(134.1740, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.2484, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(175.7782, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.6326, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.4102, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.9905, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.2772, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.7942, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(129.7074, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(145.2469, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(165.3783, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.5313, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.19148936170212766, 'recall': 0.36180904522613067, 'f1': 0.25043478260869567, 'number': 199}, 'P': {'precision': 0.14173228346456693, 'recall': 0.18181818181818182, 'f1': 0.1592920353982301, 'number': 297}, 'overall_precision': 0.166446499339498, 'overall_recall': 0.2540322580645161, 'overall_f1': 0.2011173184357542, 'overall_accuracy': 0.6601424672125691}
			------------EPOCH 22---------------
Loss:  tensor(124.5858, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.8112, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.6724, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(116.5484, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(367.5719, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(199.6657, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(119.4227, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(108.9689, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.6595, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.7653, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.3716, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.7783, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.0922, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(74.0238, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.5182, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.3737, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.2697, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.5646, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.9355, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.3461, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.9674, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(65.2994, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(87.0386, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.2412, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.4093, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.9138, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.3047, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.1309, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.6656, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.6800, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.2668, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.6950, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.7109, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(104.8721, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(62.9354, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.2316, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.3851, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.1877, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.0201, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.3329, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(151.3867, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.8132, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.5653, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.6753, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(310.2856, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(144.1352, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.3713, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(200.2263, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.1814, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.9912, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.9679, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.2333, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.6897, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(126.8514, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(144.6521, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(162.0871, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.9188, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.18024691358024691, 'recall': 0.36683417085427134, 'f1': 0.24172185430463577, 'number': 199}, 'P': {'precision': 0.17008797653958943, 'recall': 0.19528619528619529, 'f1': 0.18181818181818185, 'number': 297}, 'overall_precision': 0.17560321715817695, 'overall_recall': 0.26411290322580644, 'overall_f1': 0.21095008051529793, 'overall_accuracy': 0.6434325277944212}
			------------EPOCH 23---------------
Loss:  tensor(180.7105, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.9470, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.0533, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(108.0794, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(363.7133, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(177.8043, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(118.2188, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(97.0810, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.4731, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.5016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.1439, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.9564, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.2427, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(86.0339, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.4251, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.3857, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.1151, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(59.2050, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.1077, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.7376, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.0704, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.0968, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(87.6533, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.7934, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.9985, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.5460, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.1369, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.1896, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.0597, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.8656, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.1456, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.9740, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.9398, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(106.4711, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(62.7167, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.2996, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.3634, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.1682, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.7934, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.3923, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(174.2164, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.4743, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.7157, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.7214, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(328.9081, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(152.4689, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.3041, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(214.4366, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.9623, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.4220, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.6233, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.6339, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.7043, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(119.7629, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(129.7065, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(149.8223, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.0711, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.1978021978021978, 'recall': 0.36180904522613067, 'f1': 0.25577264653641213, 'number': 199}, 'P': {'precision': 0.1506172839506173, 'recall': 0.2053872053872054, 'f1': 0.1737891737891738, 'number': 297}, 'overall_precision': 0.1729518855656697, 'overall_recall': 0.26814516129032256, 'overall_f1': 0.21027667984189724, 'overall_accuracy': 0.6625391119099927}
			------------EPOCH 24---------------
Loss:  tensor(116.3032, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.3452, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.5798, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(103.4873, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(366.9584, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(192.5723, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(129.4208, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(104.4397, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.6038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.5114, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.2081, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.3210, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.9728, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(73.7748, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.0546, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.4723, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.8577, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.0160, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.4567, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.0167, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.6104, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.6935, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(68.1942, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.8417, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.4994, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.7423, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.5079, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.3637, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.4985, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.3707, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.2140, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.4757, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.7031, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(73.0240, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.5306, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.5837, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.0822, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.3595, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.1804, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.1365, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(133.8003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.4351, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.6801, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.8521, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(255.7981, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(114.3937, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.4922, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(164.2303, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.8135, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.7052, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.0442, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.0176, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.0467, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(105.5972, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(138.7149, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(140.5081, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.6468, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.1712962962962963, 'recall': 0.37185929648241206, 'f1': 0.23454833597464345, 'number': 199}, 'P': {'precision': 0.1620795107033639, 'recall': 0.17845117845117844, 'f1': 0.16987179487179488, 'number': 297}, 'overall_precision': 0.1673254281949934, 'overall_recall': 0.2560483870967742, 'overall_f1': 0.20239043824701194, 'overall_accuracy': 0.6436988216496904}
			------------EPOCH 25---------------
Loss:  tensor(93.1498, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.0996, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.1161, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(99.8974, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(388.4261, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(195.4568, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(160.1326, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(91.9021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.6811, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.7905, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.4286, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.8636, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.1728, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.2884, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.8572, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.9819, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.5661, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.3294, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.6988, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.9504, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.9446, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.7172, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.0028, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.4959, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.1161, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.0618, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.7571, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.7073, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.2796, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.9752, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.6566, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.4852, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.4178, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(76.8227, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.8821, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.7785, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.5442, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.3218, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.0678, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.1608, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(145.1424, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.0896, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.2894, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(308.7181, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(120.0364, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.5046, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(176.4705, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.3800, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.1835, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.5334, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5108, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.5575, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(103.1655, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(142.1056, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(141.6820, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.5082, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.17801047120418848, 'recall': 0.3417085427135678, 'f1': 0.23407917383820995, 'number': 199}, 'P': {'precision': 0.16666666666666666, 'recall': 0.22895622895622897, 'f1': 0.1929078014184397, 'number': 297}, 'overall_precision': 0.17215189873417722, 'overall_recall': 0.27419354838709675, 'overall_f1': 0.21150855365474336, 'overall_accuracy': 0.6595433060382132}
			------------EPOCH 26---------------
Loss:  tensor(90.9216, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.8472, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.6340, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(122.2841, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(368.3455, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(209.3931, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(161.5301, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(78.9667, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.8650, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.4486, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.9834, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.3831, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.4004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(71.0370, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.4297, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.7517, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.9371, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.3457, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.4771, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.4659, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.4076, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.8690, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.9976, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.9896, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.4522, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.3101, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.9777, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.1947, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.5277, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.2942, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.0450, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.2580, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(71.8806, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(145.6339, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(81.0061, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.1055, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.9877, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.0909, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.8789, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.7596, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(147.1392, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.4730, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.3890, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.1193, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(270.3629, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(121.9811, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.2595, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(167.3138, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.7132, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.5759, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.6917, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5652, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.9970, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(106.6398, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(145.1856, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(136.5774, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.3975, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.17482517482517482, 'recall': 0.3768844221105528, 'f1': 0.23885350318471338, 'number': 199}, 'P': {'precision': 0.21148036253776434, 'recall': 0.2356902356902357, 'f1': 0.2229299363057325, 'number': 297}, 'overall_precision': 0.19078947368421054, 'overall_recall': 0.2923387096774194, 'overall_f1': 0.23089171974522296, 'overall_accuracy': 0.643831968577325}
			------------EPOCH 27---------------
Loss:  tensor(80.0727, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.1961, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.6077, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(101.2030, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(328.3577, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(140.1909, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(92.9036, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(73.3538, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.4899, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.7145, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.6896, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.6104, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.2069, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(81.0401, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.7270, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.0361, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.2634, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.3105, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.8992, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.3930, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.6155, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.6809, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.5415, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.6111, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.4967, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.3409, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.5567, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.8898, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.8036, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(98.7748, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.4908, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.4201, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.5674, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(93.8286, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.0421, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.7695, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.3559, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.5831, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.7939, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.4520, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(124.3083, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.2960, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.0200, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.3238, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(241.6160, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(104.1380, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.0971, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(142.5083, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.0657, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.0091, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.2351, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.3058, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(86.4105, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(126.0899, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(108.2042, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.7541, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.17857142857142858, 'recall': 0.35175879396984927, 'f1': 0.23688663282571915, 'number': 199}, 'P': {'precision': 0.16712328767123288, 'recall': 0.2053872053872054, 'f1': 0.18429003021148035, 'number': 297}, 'overall_precision': 0.17305151915455746, 'overall_recall': 0.26411290322580644, 'overall_f1': 0.20909816440542697, 'overall_accuracy': 0.6695293256108116}
			------------EPOCH 28---------------
Loss:  tensor(87.9370, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.8140, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.6869, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(96.9792, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(310.3739, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(132.5570, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(86.7356, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(72.9058, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.8544, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.9147, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.8606, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.2240, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.2328, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.7703, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.0408, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6056, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.1380, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.9695, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.1099, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.4144, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.0434, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.1968, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.4261, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.4483, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.2968, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.0834, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.5124, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.3831, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.6011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(76.3819, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.1898, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.1351, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.7069, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(85.9504, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.4894, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.8581, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.1624, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.2535, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.8200, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.4376, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(121.1399, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.0321, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.2238, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.7480, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(227.0811, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(104.2148, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.1326, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(150.9748, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.3742, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.0716, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.9993, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.8807, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.8850, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(82.3247, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(112.0878, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(122.8938, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.9955, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.1431980906921241, 'recall': 0.3015075376884422, 'f1': 0.1941747572815534, 'number': 199}, 'P': {'precision': 0.15680473372781065, 'recall': 0.17845117845117844, 'f1': 0.16692913385826771, 'number': 297}, 'overall_precision': 0.14927344782034346, 'overall_recall': 0.22782258064516128, 'overall_f1': 0.18036711891460494, 'overall_accuracy': 0.646494907130018}
			------------EPOCH 29---------------
Loss:  tensor(86.4868, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.5297, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.7688, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(93.7257, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(313.3895, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(105.1880, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(74.7291, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(61.0322, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.8820, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.3855, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.0962, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.4079, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.7302, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.0317, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.4325, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6465, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.2651, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.6900, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.7852, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.6257, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.6005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.9656, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.8250, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.3863, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.7488, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.4073, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.3318, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2529, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.3389, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.4458, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.1481, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.1191, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.1830, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(66.3806, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.2441, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.0408, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.8350, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.5711, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.7385, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.9078, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(101.0500, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.4363, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.4954, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.4208, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(222.5866, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(94.3583, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.9485, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(170.1209, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.7891, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.6007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.0272, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3976, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.0426, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(73.1542, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(107.4586, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(104.4973, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.2358, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.17737789203084833, 'recall': 0.34673366834170855, 'f1': 0.2346938775510204, 'number': 199}, 'P': {'precision': 0.14363143631436315, 'recall': 0.17845117845117844, 'f1': 0.15915915915915918, 'number': 297}, 'overall_precision': 0.16094986807387862, 'overall_recall': 0.24596774193548387, 'overall_f1': 0.19457735247208932, 'overall_accuracy': 0.6707942214233407}
			------------EPOCH 30---------------
Loss:  tensor(146.5877, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.1453, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.8541, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(85.9286, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(280.2073, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(78.4604, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.3118, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.4550, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.4383, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.2039, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.9656, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.4477, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.6327, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.3729, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.0125, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2977, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.6995, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.9729, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.4716, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.1460, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.8924, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.7854, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.7507, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.3142, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.7467, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.9855, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.2374, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.5754, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.6830, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.0396, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.6623, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.9309, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.4801, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.7779, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.9368, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.8629, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.9261, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.8095, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5295, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.3165, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(99.9622, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.8728, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.0529, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.0689, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(213.8475, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(89.6940, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.1889, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(140.0986, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5355, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.8254, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.7025, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3875, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.9431, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(70.0778, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(97.5696, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(95.5861, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.9679, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.18952618453865336, 'recall': 0.38190954773869346, 'f1': 0.2533333333333333, 'number': 199}, 'P': {'precision': 0.20498614958448755, 'recall': 0.24915824915824916, 'f1': 0.22492401215805471, 'number': 297}, 'overall_precision': 0.1968503937007874, 'overall_recall': 0.3024193548387097, 'overall_f1': 0.23847376788553257, 'overall_accuracy': 0.6664003728113974}


		-------------RUN 3-----------
			------------EPOCH 1---------------
Loss:  tensor(2460.8691, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1416.7820, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1696.6222, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1834.4332, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1414.8328, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2334.3120, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1934.1382, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2172.6558, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1302.1655, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1311.3774, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1740.3843, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1864.0155, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2154.7976, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1413.0671, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2264.2900, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2610.0608, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2649.2002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2221.4834, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1098.0862, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1292.7245, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2107.0728, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2390.3582, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1944.5016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2041.7483, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1386.2632, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1334.1766, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1171.7870, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1674.5710, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2253.5215, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1493.5959, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2780.4968, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2609.4097, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(915.1694, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(856.5725, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1969.3995, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2314.8381, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2489.9634, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1076.1553, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1919.4244, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(944.3135, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1142.5803, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(802.9767, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1425.1525, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1374.8801, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3819.0115, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1511.0929, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2035.2485, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1763.9165, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2481.8230, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2112.5645, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1347.6033, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1855.8708, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1245.6848, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1073.7678, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(710.9286, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1052.0300, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1003.0212, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1466.7444, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.07692307692307693, 'recall': 0.022292993630573247, 'f1': 0.0345679012345679, 'number': 314}, 'P': {'precision': 0.13255360623781676, 'recall': 0.1878453038674033, 'f1': 0.15542857142857144, 'number': 362}, 'overall_precision': 0.12417218543046357, 'overall_recall': 0.11094674556213018, 'overall_f1': 0.1171875, 'overall_accuracy': 0.5550929090796097}
			------------EPOCH 2---------------
Loss:  tensor(1568.3511, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(896.6665, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1116.6439, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1194.3206, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1123.3795, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1670.5874, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1359.8547, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1840.9053, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(952.7330, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(958.4126, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1255.5305, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1278.3153, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1643.0081, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1111.6499, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1755.3418, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1970.9923, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2144.3193, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1811.5598, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(898.0427, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1054.0851, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1640.2125, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1866.5659, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1686.6733, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1833.3086, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1250.1450, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1184.1603, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(961.2964, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1518.5566, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1762.7643, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1147.0239, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2251.8867, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2011.9590, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(679.2231, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(698.6721, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1734.7225, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1836.6750, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2104.5049, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(893.9822, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1537.3278, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(797.0875, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(946.5222, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(655.5054, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1178.8744, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(963.1479, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3249.1616, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1271.6697, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1777.8274, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1489.6792, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2080.7986, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1698.5051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1082.6400, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1504.2737, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(972.7400, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(827.1747, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(573.1804, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(847.8336, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(798.4069, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1316.6479, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.08791208791208792, 'recall': 0.07643312101910828, 'f1': 0.08177172061328791, 'number': 314}, 'P': {'precision': 0.1686746987951807, 'recall': 0.19337016574585636, 'f1': 0.1801801801801802, 'number': 362}, 'overall_precision': 0.13662790697674418, 'overall_recall': 0.1390532544378698, 'overall_f1': 0.13782991202346043, 'overall_accuracy': 0.6113976757193462}
			------------EPOCH 3---------------
Loss:  tensor(1340.6226, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(721.8234, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(940.9535, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(985.5625, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(949.4531, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1446.3901, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1146.2017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1672.3036, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(721.1321, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(767.8931, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1038.7456, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1001.2899, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1312.3925, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(939.8250, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1443.9895, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1656.2683, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1803.0038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1556.3032, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(769.1993, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(888.3804, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1286.9723, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1399.7561, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1380.7993, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1591.5811, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(810.9147, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(789.9056, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(729.2739, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1131.3317, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1510.9604, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(917.3853, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1921.6869, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1759.6565, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(549.4823, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(603.5773, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1317.9811, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1569.3594, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1728.5505, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(736.9474, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1189.9836, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(681.2783, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(733.7295, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(531.4128, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(930.8220, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(705.7917, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2730.5039, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(998.9299, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1485.9321, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1261.1239, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1831.6790, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1376.2651, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(809.1555, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1186.2601, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(772.3535, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(684.0976, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(458.4418, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(719.6992, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(653.0105, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1161.8130, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.15473441108545036, 'recall': 0.21337579617834396, 'f1': 0.17938420348058906, 'number': 314}, 'P': {'precision': 0.25098039215686274, 'recall': 0.17679558011049723, 'f1': 0.20745542949756887, 'number': 362}, 'overall_precision': 0.19040697674418605, 'overall_recall': 0.1937869822485207, 'overall_f1': 0.19208211143695014, 'overall_accuracy': 0.5996519793673482}
			------------EPOCH 4---------------
Loss:  tensor(1206.6177, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(570.9560, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(775.9570, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(804.3870, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(813.4994, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1320.1000, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1061.7211, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1399.8594, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(605.7876, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(653.1073, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(912.9528, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(808.1775, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1089.0056, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(780.5057, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1193.4916, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1409.9246, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1571.0048, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1371.8071, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(660.2216, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(741., device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1117.1066, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1047.5189, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1151.5526, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1495.0889, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(551.4253, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(563.4235, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(570.7310, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(938.3845, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1271.8491, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(691.1628, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1697.2291, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1462.7522, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(422.3202, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(503.7659, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(944.7640, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1292.5786, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1432., device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(596.4493, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(995.7841, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(574.0809, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(561.9683, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(429.3622, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(765.9279, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(530.1248, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2472.0647, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(864.2907, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1268.2748, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(991.1451, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1644.4807, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1128.2734, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(611.8271, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(962.3446, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(586.1938, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(539.3730, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(351.6264, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(628.9257, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(553.5081, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1055.8215, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.24173027989821882, 'recall': 0.30254777070063693, 'f1': 0.2687411598302687, 'number': 314}, 'P': {'precision': 0.2692307692307692, 'recall': 0.27071823204419887, 'f1': 0.26997245179063356, 'number': 362}, 'overall_precision': 0.25495376486129456, 'overall_recall': 0.28550295857988167, 'overall_f1': 0.2693649685973482, 'overall_accuracy': 0.635324094214157}
			------------EPOCH 5---------------
Loss:  tensor(1071.3079, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(444.9207, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(599.8364, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(680.0782, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(715.7703, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1132.5034, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(936.2279, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1286.5332, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(573.5132, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(558.2919, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(830.8106, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(677.9673, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(990.6818, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(739.9766, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1039.2219, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1442.5862, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1567.4313, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1382.2085, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(567.0154, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(632.2841, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(984.5726, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(909.8014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(991.7233, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1274.4824, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(388.8615, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(414.3579, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(433.2562, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(717.9888, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(997.4781, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(559.7624, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1443.6963, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1380.9857, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(386.2490, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(446.9933, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(802.2367, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1169.9958, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1315.0900, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(569.9656, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(877.0936, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(483.6367, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(492.2416, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(440.7459, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(671.6762, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(436.4339, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2160.4236, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(699.6573, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1087.7546, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(900.1838, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1374.2517, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(930.1956, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(530.3156, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(734.1973, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(407.4282, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(405.3687, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(255.6978, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(472.1344, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(587.4197, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1020.1437, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.18719211822660098, 'recall': 0.12101910828025478, 'f1': 0.1470019342359768, 'number': 314}, 'P': {'precision': 0.2434108527131783, 'recall': 0.43370165745856354, 'f1': 0.3118172790466733, 'number': 362}, 'overall_precision': 0.22995283018867924, 'overall_recall': 0.28846153846153844, 'overall_f1': 0.2559055118110236, 'overall_accuracy': 0.5833074389410229}
			------------EPOCH 6---------------
Loss:  tensor(1329.1821, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(545.7552, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(601.2300, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(964.2600, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(683.6880, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(884.5032, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(924.5084, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1250.4094, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(448.2186, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(411.2979, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(743.8148, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(560.2522, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(999.4641, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(765.2371, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(934.6365, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1725.5773, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1604.9718, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1577.7256, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(632.5015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(846.7352, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1369.6913, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1217.7467, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1129.4016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1350.6650, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(527.8066, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(447.5229, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(337.1130, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(730.9385, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(863.2390, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(518.1993, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1225.8606, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1317.3665, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(315.8287, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(314.8964, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(882.9623, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1012.4001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1288.3860, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(473.3101, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(843.1740, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(525.1473, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(441.5352, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(378.8592, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(690.8141, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(378.7303, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2517.3208, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(747.0436, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1139.7175, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1011.9390, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1658.2097, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1104.8059, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(682.6851, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(889.7297, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(569.9531, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(504.6794, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(350.3033, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(390.4424, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(527.5330, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1120.3318, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.22477064220183487, 'recall': 0.15605095541401273, 'f1': 0.18421052631578946, 'number': 314}, 'P': {'precision': 0.12849162011173185, 'recall': 0.1270718232044199, 'f1': 0.12777777777777777, 'number': 362}, 'overall_precision': 0.16493055555555555, 'overall_recall': 0.14053254437869822, 'overall_f1': 0.15175718849840256, 'overall_accuracy': 0.6450189546951712}
			------------EPOCH 7---------------
Loss:  tensor(1142.3120, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(511.5271, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(648.9056, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(721.7192, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(580.5378, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(734.8775, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(617.5757, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(845.6064, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(343.9416, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(341.5333, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(612.2382, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(505.5872, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(794.2949, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(627.1635, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(927.0206, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1075.8674, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1268.2200, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1116.2334, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(476.9543, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(507.8934, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(951.3961, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(827.6960, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(847.3892, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1204.3137, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(423.3195, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(324.0565, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(232.2337, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(610.9384, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1114.7649, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(579.7433, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1404.9446, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1698.4585, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(437.3159, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(262.5814, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1202.8546, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1111.5809, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(974.7183, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(338.5130, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(671.5981, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(417.6864, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(353.5626, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(294.8605, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(542.7745, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(506.6228, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1569.8892, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(467.4136, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(730.9456, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(607.5535, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1185.4639, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(733.9829, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(355.4695, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(574.1506, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(274.5174, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(278.1620, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(167.7049, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(241.9311, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(361.2558, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(627.0689, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.13978494623655913, 'recall': 0.08280254777070063, 'f1': 0.104, 'number': 314}, 'P': {'precision': 0.11809045226130653, 'recall': 0.1298342541436464, 'f1': 0.12368421052631579, 'number': 362}, 'overall_precision': 0.125, 'overall_recall': 0.10798816568047337, 'overall_f1': 0.11587301587301588, 'overall_accuracy': 0.6060530731464794}
			------------EPOCH 8---------------
Loss:  tensor(1068.5975, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(330.3329, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(461.1691, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(591.3289, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(569.8888, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(615.5557, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(671.8550, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(803.1359, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(368.4250, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(335.6860, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(583.0284, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(538.7413, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(776.5226, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(601.4255, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1073.2252, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1167.5004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1030.7059, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(871.8840, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(481.7754, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(444.4363, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(822.9936, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(686.3579, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(668.2301, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(809.7015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(213.0163, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(204.3666, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(192.6591, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(409.2784, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(527.8931, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(280.4183, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(772.0446, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(987.7078, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(303.0877, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(204.4652, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(596.8083, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(704.7909, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(741.7720, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(279.9828, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(580.0307, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(348.1595, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(366.3463, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(252.2292, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(544.5082, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(476.5733, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1602.3636, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(449.3098, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(625.7236, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(462.1550, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1095.2559, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(745.6765, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(402.7959, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(540.5203, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(296.1648, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(267.3148, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(151.2469, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(271.6003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(286.9910, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(616.4111, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.18777292576419213, 'recall': 0.13694267515923567, 'f1': 0.15837937384898712, 'number': 314}, 'P': {'precision': 0.22167487684729065, 'recall': 0.3729281767955801, 'f1': 0.2780638516992791, 'number': 362}, 'overall_precision': 0.21241050119331742, 'overall_recall': 0.26331360946745563, 'overall_f1': 0.23513870541611628, 'overall_accuracy': 0.6264371387732273}
			------------EPOCH 9---------------
Loss:  tensor(802.2034, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(214.5853, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(303.0197, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(413.8197, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(470.6517, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(499.0536, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(527.9791, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(656.9355, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(221.4908, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(212.1110, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(510.9496, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(489.9908, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(574.5423, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(425.0633, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(656.5340, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1099.9061, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(869.4707, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(739.6146, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(354.6774, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(286.7997, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(772.7397, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(649.0492, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(544.8019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(720.4323, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(202.2818, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(217.3621, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(143.7976, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(351.4898, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(604.9680, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(286.8567, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(840.0012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1041.0059, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(283.2766, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(235.2097, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(404.9244, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(638.9595, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(564.5966, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(254.2943, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(389.2953, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(247.9056, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(182.2744, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(191.2215, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(335.5772, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(230.7657, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1085.1240, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(298.6884, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(475.9851, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(308.0472, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(790.6135, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(500.3227, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(245.2132, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(386.4136, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(193.2726, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(174.6063, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(100.4232, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(136.9782, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(207.4136, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(478.1334, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.27167630057803466, 'recall': 0.14968152866242038, 'f1': 0.19301848049281312, 'number': 314}, 'P': {'precision': 0.20973782771535582, 'recall': 0.30939226519337015, 'f1': 0.25000000000000006, 'number': 362}, 'overall_precision': 0.2248939179632249, 'overall_recall': 0.23520710059171598, 'overall_f1': 0.2299349240780911, 'overall_accuracy': 0.6358212665465167}
			------------EPOCH 10---------------
Loss:  tensor(733.2455, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(158.4072, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(246.1105, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(330.2152, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(317.6148, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(351.1932, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(364.6818, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(475.8096, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(195.4973, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(161.7017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(379.3239, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(285.8059, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(469.4285, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(361.2773, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(541.0098, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(711.9294, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(697.6327, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(576.5453, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(307.7515, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(214.6207, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(501.7401, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(448.6137, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(419.1310, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(542.4270, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(112.8564, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(120.5546, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(78.1695, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(255.3509, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(332.0451, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(186.1446, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(444.9238, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(695.1133, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(204.3546, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(160.1032, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(305.6828, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(350.3290, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(387.8713, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(216.2209, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(334.4250, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(196.0387, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(211.9511, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(187.1391, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(362.6566, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(277.6850, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1205.3805, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(376.8256, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(554.8114, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(425.7992, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(886.0278, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(538.9636, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(297.9735, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(461.5766, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(164.9185, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(137.4276, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(128.5102, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(110.2017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(193.9417, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(337.2556, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.27314814814814814, 'recall': 0.18789808917197454, 'f1': 0.22264150943396227, 'number': 314}, 'P': {'precision': 0.1887966804979253, 'recall': 0.2513812154696133, 'f1': 0.2156398104265403, 'number': 362}, 'overall_precision': 0.2148997134670487, 'overall_recall': 0.22189349112426035, 'overall_f1': 0.21834061135371177, 'overall_accuracy': 0.663973649866385}
			------------EPOCH 11---------------
Loss:  tensor(542.2972, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(121.3768, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(178.1997, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(224.7174, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(239.3089, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(240.3998, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(303.1910, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(357.8462, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(176.4933, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(145.2052, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(325.8373, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(237.0858, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(619.4242, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(476.3259, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(780.2768, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(771.6251, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(726.1201, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(646.2603, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(395.2933, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(273.4028, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(608.8937, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(447.0217, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(381.3138, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(546.8953, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(104.5629, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(105.6429, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(86.4859, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(256.4553, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(261.3606, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(147.2834, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(383.7589, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(567.3165, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(151.2554, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(134.8344, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(264.4323, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(255.7370, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(350.4091, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(147.4591, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(319.8881, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(163.9540, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(119.2294, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(128.3896, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(285.5271, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(273.4612, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1119.2875, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(290.0222, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(451.5141, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(319.2749, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(814.4460, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(490.4654, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(310.3613, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(499.1238, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(182.5013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(135.6184, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(150.4569, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(148.4047, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(259.3713, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(348.8838, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.23515439429928742, 'recall': 0.31528662420382164, 'f1': 0.2693877551020408, 'number': 314}, 'P': {'precision': 0.1423728813559322, 'recall': 0.11602209944751381, 'f1': 0.1278538812785388, 'number': 362}, 'overall_precision': 0.19692737430167598, 'overall_recall': 0.20857988165680474, 'overall_f1': 0.20258620689655174, 'overall_accuracy': 0.6614877882045864}
			------------EPOCH 12---------------
Loss:  tensor(596.7851, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(182.6828, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(211.6399, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(262.1987, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(247.3574, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(216.5313, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(276.2958, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(307.4270, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(132.8599, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(99.8883, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(224.7527, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(210.2277, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(387.5710, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(293.3400, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(504.9113, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(576.0876, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(407.4854, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(396.2718, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(215.7890, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(143.7059, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(456.5902, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(350.9277, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(274.7986, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(436.1366, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(156.8914, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(171.0273, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(205.1556, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(583.6815, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(663.2811, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(461.6083, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(959.8115, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(808.4048, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(181.9335, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(216.9857, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(623.6370, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(507.6306, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(459.0916, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(130.7762, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(233.2271, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(201.3921, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.3058, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(86.6171, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(243.0100, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(114.9242, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1043.0859, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(234.5862, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(393.5575, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(182.8574, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(619.2830, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(403.2072, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(274.3815, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(451.2388, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(214.8883, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(142.8858, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(113.3975, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(164.8371, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(394.8267, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(362.7353, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.21372549019607842, 'recall': 0.3471337579617834, 'f1': 0.2645631067961165, 'number': 314}, 'P': {'precision': 0.1476510067114094, 'recall': 0.06077348066298342, 'f1': 0.08610567514677103, 'number': 362}, 'overall_precision': 0.19878603945371776, 'overall_recall': 0.1937869822485207, 'overall_f1': 0.19625468164794005, 'overall_accuracy': 0.6122677273009757}
			------------EPOCH 13---------------
Loss:  tensor(602.6440, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(281.4607, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(385.7011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(335.5999, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(305.7769, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(694.8539, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(495.4737, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(475.8903, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(363.1861, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(283.1338, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(361.2913, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(256.1303, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(449.2997, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(315.9148, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(528.9914, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(662.8765, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(505.9895, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(509.9450, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(268.8922, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(218.2933, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(434.3518, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(413.9147, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(326.3833, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(381.4229, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(91.2090, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(78.3746, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(71.8525, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(178.2032, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(256.8870, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(123.3370, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(355.0359, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(532.6071, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(106.5333, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(122.5389, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(303.8144, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(313.9189, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(430.2022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(210.0197, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(337.9743, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(288.2454, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(130.9981, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(215.2660, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(416.8105, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(225.0869, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1422.5240, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(281.8127, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(553.8894, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(540.3932, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1037.2273, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(527.9100, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(233.7128, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(400.1163, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(177.5800, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(121.0043, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(88.1200, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(107.1564, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(167.0405, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(272.4745, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.2787878787878788, 'recall': 0.2929936305732484, 'f1': 0.28571428571428575, 'number': 314}, 'P': {'precision': 0.21428571428571427, 'recall': 0.26519337016574585, 'f1': 0.23703703703703702, 'number': 362}, 'overall_precision': 0.2416452442159383, 'overall_recall': 0.2781065088757396, 'overall_f1': 0.2585969738651994, 'overall_accuracy': 0.6804424833758002}
			------------EPOCH 14---------------
Loss:  tensor(459.8429, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(95.4109, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(155.3108, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(192.3022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(154.2155, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(175.2463, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(207.2802, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(254.2106, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(131.7762, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(89.8184, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(193.6953, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(159.3888, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(287.6762, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(218.2708, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(351.8392, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(478.9674, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(392.5621, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(399.0625, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(218.6109, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(177.3364, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(493.6932, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(487.7509, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(297.3813, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(326.8254, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(107.6040, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(81.1109, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.7548, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(139.8126, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(421.8818, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(179.0055, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(408.9041, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(730.1620, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(156.0066, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(154.6933, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(192.6290, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(357.8647, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(305.5945, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(116.3623, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(246.7343, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(113.7753, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.6109, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(68.6522, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(166.5243, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(121.8330, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(656.6771, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(134.0038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(228.0535, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(126.9098, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(445.1256, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(227.8099, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(115.7034, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(201.8399, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(97.2073, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(71.9028, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.6670, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.5881, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(95.3293, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(171.9732, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.29880478087649404, 'recall': 0.23885350318471338, 'f1': 0.26548672566371684, 'number': 314}, 'P': {'precision': 0.21181262729124237, 'recall': 0.287292817679558, 'f1': 0.24384525205158264, 'number': 362}, 'overall_precision': 0.24123989218328842, 'overall_recall': 0.26479289940828404, 'overall_f1': 0.2524682651622003, 'overall_accuracy': 0.689142999192095}
			------------EPOCH 15---------------
Loss:  tensor(416.8111, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(80.1750, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(127.4642, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(134.2870, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(134.4239, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(91.2900, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(165.5941, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(192.6578, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(101.9859, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(72.0552, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(116.4558, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(114.2258, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(241.3018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(216.7878, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(370.8012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(404.5182, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(259.4498, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(274.9941, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(135.3642, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(72.3091, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(325.2033, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(326.9285, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(236.7285, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(262.8734, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.4497, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.7292, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.8782, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(136.6060, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(165.3817, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(88.1909, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(183.8382, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(413.5898, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.1715, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(73.7516, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(139.7992, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(74.2312, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(198.7656, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.6083, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(130.3287, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(94.1758, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.4092, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.3758, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(129.8338, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(72.3126, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(566.1937, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(107.9290, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(166.1225, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(86.9021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(355.7885, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(162.8701, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(87.8853, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(159.8830, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(75.5604, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.8324, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.2055, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.4117, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(64.7795, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(123.9152, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.2620481927710843, 'recall': 0.2770700636942675, 'f1': 0.2693498452012384, 'number': 314}, 'P': {'precision': 0.16, 'recall': 0.17679558011049723, 'f1': 0.1679790026246719, 'number': 362}, 'overall_precision': 0.2062841530054645, 'overall_recall': 0.22337278106508876, 'overall_f1': 0.21448863636363635, 'overall_accuracy': 0.6954819464296812}
			------------EPOCH 16---------------
Loss:  tensor(300.9389, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.3533, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(69.0492, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(96.6404, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(65.2950, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.4800, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(136.5299, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(118.2269, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(67.6368, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.4179, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(88.2462, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(93.5239, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(196.5927, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(158.4888, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(293.2571, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(346.3326, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(300.0973, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(273.4270, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(96.8400, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.1419, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(288.4053, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(222.3934, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(121.1272, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(190.2450, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.6634, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.0548, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.8209, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(68.8980, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(158.1198, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(67.3278, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(151.0056, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(360.2413, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.9840, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(69.7760, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(89.8016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.6771, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(155.6893, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.8093, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(102.9116, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(70.5825, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.5244, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.8406, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(103.8467, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.6305, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(557.6530, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(94.4429, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(146.0266, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(86.2980, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(329.6878, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(167.0984, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(85.9820, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(146.2517, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(85.1176, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.7659, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.5826, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.9735, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.6989, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(93.4115, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.25, 'recall': 0.2802547770700637, 'f1': 0.2642642642642643, 'number': 314}, 'P': {'precision': 0.2102803738317757, 'recall': 0.24861878453038674, 'f1': 0.2278481012658228, 'number': 362}, 'overall_precision': 0.2282051282051282, 'overall_recall': 0.26331360946745563, 'overall_f1': 0.24450549450549453, 'overall_accuracy': 0.6900130507737244}
			------------EPOCH 17---------------
Loss:  tensor(245.9717, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.3253, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.8717, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(86.8396, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.0628, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.3724, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(106.3424, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(79.7221, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.2727, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.6663, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(66.4254, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(59.6534, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(128.8400, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(111.3608, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(210.2121, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(315.9172, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(310.2421, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(246.8679, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(79.7193, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.3321, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(234.2209, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(207.8412, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(126.4097, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(165.9648, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.2817, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.4175, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.1563, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(80.1541, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(109.0763, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.5511, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(137.2900, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(352.6722, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.7889, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.7275, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(68.6581, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.0616, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(144.2446, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.6076, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(92.2416, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.2979, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.6167, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.9231, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(76.4494, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.6134, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(483.5092, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(82.2019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(116.7568, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.9606, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(269.0116, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(121.3321, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(62.9949, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(114.3023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.7831, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.2298, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.6234, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.3130, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.9105, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(84.3281, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.29069767441860467, 'recall': 0.3184713375796178, 'f1': 0.303951367781155, 'number': 314}, 'P': {'precision': 0.23604060913705585, 'recall': 0.2569060773480663, 'f1': 0.24603174603174602, 'number': 362}, 'overall_precision': 0.26151761517615174, 'overall_recall': 0.28550295857988167, 'overall_f1': 0.272984441301273, 'overall_accuracy': 0.6842334224100429}
			------------EPOCH 18---------------
Loss:  tensor(222.6229, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.2664, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.8083, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(75.7649, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(61.0744, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.2237, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(113.6120, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.3947, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.5684, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.6028, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.1281, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.5877, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(120.2781, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(103.4195, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(193.4320, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(247.7034, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(194.7055, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(167.3682, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(62.7975, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.7145, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(190.2182, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(171.8100, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(98.7457, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(127.8806, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.1628, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.8279, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.2681, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.1886, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(86.8692, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.3060, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(100.9515, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(305.1270, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.6649, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.6539, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.7823, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.8113, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(142.1702, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.7339, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(91.9031, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.3778, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.9598, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.6437, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(68.2914, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.8213, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(417.2189, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(71.7377, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(101.7919, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.2978, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(241.4754, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(92.8888, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.9039, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(104.2110, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.8994, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.4555, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.4735, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.8913, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.8235, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.8404, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.29780564263322884, 'recall': 0.30254777070063693, 'f1': 0.3001579778830964, 'number': 314}, 'P': {'precision': 0.21291866028708134, 'recall': 0.24585635359116023, 'f1': 0.2282051282051282, 'number': 362}, 'overall_precision': 0.2496607869742198, 'overall_recall': 0.27218934911242604, 'overall_f1': 0.26043878273177634, 'overall_accuracy': 0.6905723696476291}
			------------EPOCH 19---------------
Loss:  tensor(175.7751, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.4314, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.2522, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.9005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.2244, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.7094, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(92.4700, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.9232, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.4053, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.4750, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.0272, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.7233, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(96.3660, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(78.3396, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(158.4254, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(228.9197, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(156.4999, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(143.6701, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.7799, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.1373, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(171.5977, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(176.2435, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(112.8491, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(120.7706, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.1348, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.0173, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.9434, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.1342, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(80.1079, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.6069, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(86.2744, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(274.4143, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.2244, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.5564, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.3353, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.8003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(119.7580, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.1864, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(71.7547, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.6813, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.3622, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.1840, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.0955, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.3779, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(366.9010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.5893, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(83.0923, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.2430, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(211.6077, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(76.1748, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.5597, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(89.5126, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.9743, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.9470, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.4580, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.4586, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.4075, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.7371, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.2838709677419355, 'recall': 0.2802547770700637, 'f1': 0.28205128205128205, 'number': 314}, 'P': {'precision': 0.1686746987951807, 'recall': 0.19337016574585636, 'f1': 0.1801801801801802, 'number': 362}, 'overall_precision': 0.21793103448275863, 'overall_recall': 0.23372781065088757, 'overall_f1': 0.225553176302641, 'overall_accuracy': 0.6880243614442856}
			------------EPOCH 20---------------
Loss:  tensor(161.7765, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.1563, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.0007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.2300, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.3209, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.4204, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(78.7841, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.7879, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.1842, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.7972, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.0882, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.2388, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(70.8120, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(61.5284, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(127.8530, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(231.1084, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(142.7612, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(125.8092, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.3610, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.2518, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(148.6541, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(158.1341, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(98.3477, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(90.4790, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.1501, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.6408, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.7153, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.1679, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(69.5182, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.8398, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(78.4555, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(251.8859, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.6914, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.4154, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.4205, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.8886, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(120.2620, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.1944, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(66.3084, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.1342, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.7738, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.2096, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.1396, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.7872, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(416.4775, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(69.0477, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(88.4678, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.0570, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(215.5692, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(76.2817, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.5753, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(72.1377, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.2995, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.4857, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.7887, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.6021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.7886, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.0985, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.2721518987341772, 'recall': 0.27388535031847133, 'f1': 0.273015873015873, 'number': 314}, 'P': {'precision': 0.17016317016317017, 'recall': 0.20165745856353592, 'f1': 0.18457648546144126, 'number': 362}, 'overall_precision': 0.2134228187919463, 'overall_recall': 0.23520710059171598, 'overall_f1': 0.22378606615059818, 'overall_accuracy': 0.6864085513641166}
			------------EPOCH 21---------------
Loss:  tensor(140.2900, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.5570, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.5413, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.5694, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.8435, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.0899, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(73.8217, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.3387, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.7262, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.5300, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.0246, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.7366, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.8932, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.4978, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(104.2661, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(217.1016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(132.9345, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(116.5502, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.5946, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.0000, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(157.7664, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(155.5496, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(92.0525, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(85.1753, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.3189, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.7767, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.0869, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.6750, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.7705, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.2800, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.4617, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(258.3034, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.1690, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.0658, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.4427, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.9770, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(111.2644, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.0635, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.3891, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.0247, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.3744, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.1723, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.9323, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.9405, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(364.0454, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.9696, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(77.3541, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.0326, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(184.9920, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(64.6186, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.8330, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(71.2728, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.9893, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.9487, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.6087, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.7659, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.0633, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.4506, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.2708860759493671, 'recall': 0.34076433121019106, 'f1': 0.30183356840620595, 'number': 314}, 'P': {'precision': 0.232, 'recall': 0.24033149171270718, 'f1': 0.23609226594301222, 'number': 362}, 'overall_precision': 0.2519480519480519, 'overall_recall': 0.2869822485207101, 'overall_f1': 0.26832641770401106, 'overall_accuracy': 0.6825554657883289}
			------------EPOCH 22---------------
Loss:  tensor(164.2031, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.3018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.3884, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.5971, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.3286, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.7744, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(140.3928, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.2286, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.5794, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.1295, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.1964, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.2064, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(73.5193, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(70.7257, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(147.6416, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(213.9022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(164.9485, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(122.2990, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.9080, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.9438, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(255.6725, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(183.8804, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(61.5391, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(79.2256, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.2363, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.0754, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.6883, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.9220, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(72.7986, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.9929, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(69.3645, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(285.0101, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.7005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.6510, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.4810, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.6578, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(111.8880, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.9540, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.0879, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.8392, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.6073, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.5741, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.2870, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.1296, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(411.1331, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.6438, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(70.4003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.4668, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(204.8428, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.0122, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.7946, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(79.0227, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.0269, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.4234, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.4908, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.2705, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.2756, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.7589, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.28700906344410876, 'recall': 0.30254777070063693, 'f1': 0.29457364341085274, 'number': 314}, 'P': {'precision': 0.22966507177033493, 'recall': 0.26519337016574585, 'f1': 0.24615384615384617, 'number': 362}, 'overall_precision': 0.25500667556742324, 'overall_recall': 0.28254437869822485, 'overall_f1': 0.26807017543859646, 'overall_accuracy': 0.6869678702380213}
			------------EPOCH 23---------------
Loss:  tensor(166.2803, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.4016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.7300, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.2024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.0706, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.3912, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(118.2086, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.4539, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.2989, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.6003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.6977, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.5752, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(99.3087, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(89.6339, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(187.8735, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(191.4077, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(128.9919, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(115.9939, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.0095, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.9269, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(159.6092, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(129.0556, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.2973, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(76.8328, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.7255, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.5576, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.8921, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.3799, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.5501, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.4575, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.6094, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(232.5803, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.2619, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.0540, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.8455, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.8918, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(115.2031, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.8226, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(83.9034, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.4534, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.7276, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.7636, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.6851, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.2172, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(425.7253, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(71.3755, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(92.9700, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.3381, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(201.0694, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.7878, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.9479, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(65.3921, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.4969, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.4269, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.1219, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.2619, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.4559, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.3837, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.26256983240223464, 'recall': 0.29936305732484075, 'f1': 0.27976190476190477, 'number': 314}, 'P': {'precision': 0.21994884910485935, 'recall': 0.23756906077348067, 'f1': 0.22841965471447545, 'number': 362}, 'overall_precision': 0.24032042723631508, 'overall_recall': 0.26627218934911245, 'overall_f1': 0.25263157894736843, 'overall_accuracy': 0.6884593872351004}
			------------EPOCH 24---------------
Loss:  tensor(117.2255, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.4670, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.9679, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.3267, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.6526, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.3145, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(72.6347, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.4310, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.6602, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.8547, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.5816, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.6893, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(79.1041, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(77.0910, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(158.8638, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(167.7256, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(124.5243, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(102.8633, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.4968, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.6759, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(126.9854, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(106.8644, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.6369, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(73.7533, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.3238, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.2631, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.7374, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.1972, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.8312, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.7266, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.4155, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(247.4053, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.0191, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.5710, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.2835, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.7826, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(94.5769, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.1510, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.0076, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.0059, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.9691, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.7965, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.9446, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.9395, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(297.6189, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.6103, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.1174, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.7526, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(144.0212, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.2776, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.3155, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.0857, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.4747, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.2161, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.4399, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.3725, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.4083, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.5424, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.27245508982035926, 'recall': 0.2898089171974522, 'f1': 0.28086419753086417, 'number': 314}, 'P': {'precision': 0.17142857142857143, 'recall': 0.19889502762430938, 'f1': 0.18414322250639387, 'number': 362}, 'overall_precision': 0.21618037135278514, 'overall_recall': 0.2411242603550296, 'overall_f1': 0.22797202797202795, 'overall_accuracy': 0.68908085265055}
			------------EPOCH 25---------------
Loss:  tensor(113.7450, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.1559, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.0090, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.8782, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.1280, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.1488, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(67.0850, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.6029, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.0130, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.6334, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.3844, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.7678, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(75.6932, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(74.1228, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(150.4427, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(160.1268, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(133.4045, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(98.2207, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.6682, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.3220, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(142.2891, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(110.8646, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.4217, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(69.8088, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.5649, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.4507, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.1958, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.6822, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.2828, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.8878, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.5283, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(218.6405, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.3820, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.7872, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.3299, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.2167, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(87.9862, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.5647, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.1249, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.9702, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.8884, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.2522, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.4479, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.4637, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(316.1384, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.2688, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(61.8191, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.4230, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(143.5916, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.7722, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.0591, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.6651, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.1040, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.4395, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.7387, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.7080, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.4111, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.4029, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.27325581395348836, 'recall': 0.29936305732484075, 'f1': 0.2857142857142857, 'number': 314}, 'P': {'precision': 0.1920374707259953, 'recall': 0.2265193370165746, 'f1': 0.2078580481622307, 'number': 362}, 'overall_precision': 0.22827496757457846, 'overall_recall': 0.2603550295857988, 'overall_f1': 0.2432619212163096, 'overall_accuracy': 0.6932446709340625}
			------------EPOCH 26---------------
Loss:  tensor(96.0131, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.5203, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.9957, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.1939, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.9687, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.5570, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.2592, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.0131, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.8197, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.4860, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.4038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.4294, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.2304, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.9385, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(73.4782, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(161.7889, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(122.0793, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(91.4499, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.3799, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.3560, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(128.4330, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(104.8840, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.3771, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(67.7607, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.5668, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.1600, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.7545, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.6295, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.3064, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.8347, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.6528, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(186.6090, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.8410, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.2985, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.8202, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.6262, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(85.2326, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.1192, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.8328, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.5006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.1814, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.7514, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.4165, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.3757, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(232.5669, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.7067, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.1505, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.9315, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(121.3465, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.7964, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.8934, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.8781, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.0727, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.8880, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.1140, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5446, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.5278, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.7586, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.2739018087855297, 'recall': 0.3375796178343949, 'f1': 0.30242510699001424, 'number': 314}, 'P': {'precision': 0.20099255583126552, 'recall': 0.22375690607734808, 'f1': 0.21176470588235294, 'number': 362}, 'overall_precision': 0.23670886075949368, 'overall_recall': 0.27662721893491127, 'overall_f1': 0.2551159618008185, 'overall_accuracy': 0.6842955689515878}
			------------EPOCH 27---------------
Loss:  tensor(92.4590, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.5170, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.1706, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.1866, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.3616, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.8125, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.9778, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.5995, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.4538, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.8884, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.3018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.0313, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.9689, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.8768, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(67.3861, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(164.3676, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(98.1894, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(80.9816, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.0038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.9513, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(116.9583, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(136.2286, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(79.8391, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.9555, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.0786, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.8643, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5300, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.5103, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.4612, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.3299, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.5706, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(170.0484, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.2163, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.9758, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.9311, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.0602, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(80.0774, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.6355, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.7042, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.4801, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.1950, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.6673, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.8460, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.1979, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(251.7640, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.4935, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.2394, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.7579, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(113.8681, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.0079, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.3077, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.4582, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.3158, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.5938, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.6493, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.6362, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.6823, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.4775, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.27575757575757576, 'recall': 0.2898089171974522, 'f1': 0.28260869565217395, 'number': 314}, 'P': {'precision': 0.20763723150357996, 'recall': 0.24033149171270718, 'f1': 0.22279129321382846, 'number': 362}, 'overall_precision': 0.2376502002670227, 'overall_recall': 0.26331360946745563, 'overall_f1': 0.24982456140350878, 'overall_accuracy': 0.695730532595861}
			------------EPOCH 28---------------
Loss:  tensor(92.3672, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.2411, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.7291, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.0338, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.7731, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.1728, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.9982, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.9734, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.5660, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.1820, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.7650, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.3731, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.6802, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.3359, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.0970, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(141.0573, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(84.8887, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(70.8727, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.6493, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.1958, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(86.6400, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(105.4815, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(66.3722, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.5240, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.4428, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.0011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.3047, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.0911, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.3025, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.8078, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.0764, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(175.8271, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.5878, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.6585, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.5469, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.1304, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(88.5168, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.9401, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.5306, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.5663, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4338, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.8620, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.9627, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.1398, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(300.3871, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.9748, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.7191, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.0012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(120.1717, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.7235, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.0793, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.3826, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.8903, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.7934, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.7309, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.0991, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.4596, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.0934, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.25914634146341464, 'recall': 0.27070063694267515, 'f1': 0.264797507788162, 'number': 314}, 'P': {'precision': 0.16997792494481237, 'recall': 0.212707182320442, 'f1': 0.18895705521472395, 'number': 362}, 'overall_precision': 0.20742637644046094, 'overall_recall': 0.23964497041420119, 'overall_f1': 0.22237474262182566, 'overall_accuracy': 0.6892672922751849}
			------------EPOCH 29---------------
Loss:  tensor(74.9158, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.8700, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.2384, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.1985, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.7865, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.3161, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.6198, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.4493, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.0267, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.6845, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.4909, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.4674, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.4057, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.8312, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.8770, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(179.3486, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(88.6097, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(78.3202, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.3245, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.9958, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(115.1931, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(97.7543, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.3079, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.1716, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.3123, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.0270, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.9006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.1453, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.2048, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.1551, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.9713, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(191.6110, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.7015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.9910, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.3951, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.8244, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(87.9135, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.7092, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.9043, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.2248, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1682, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.4935, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.3726, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.2119, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(206.9153, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.0042, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.1567, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.7868, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(104.7287, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.9847, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.4957, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.1560, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.2312, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.9310, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.6498, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2422, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.3641, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.2901, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.2597402597402597, 'recall': 0.3184713375796178, 'f1': 0.28612303290414876, 'number': 314}, 'P': {'precision': 0.23097112860892388, 'recall': 0.2430939226519337, 'f1': 0.23687752355316286, 'number': 362}, 'overall_precision': 0.2454308093994778, 'overall_recall': 0.2781065088757396, 'overall_f1': 0.260748959778086, 'overall_accuracy': 0.6820582934559691}
			------------EPOCH 30---------------
Loss:  tensor(79.3386, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.4927, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.1852, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.5783, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.6354, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.4693, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.8476, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.9901, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.9217, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.7176, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.8174, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.7068, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.0731, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.2168, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(59.2169, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(124.8235, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(91.4365, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(76.4933, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.6598, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5709, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(186.0239, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(124.9749, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.0334, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.5429, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.3315, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.7092, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4990, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.0843, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.2337, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.5591, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.2117, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(156.4445, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5121, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.8205, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.5815, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.7934, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(72.1290, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.9454, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.4128, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.5018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6871, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.6694, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.3889, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.7011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(182.5505, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.8307, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.2441, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.6996, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(101.5311, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.1729, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.4180, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.1847, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.9528, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.0518, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.2953, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2482, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.6120, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.5246, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.29245283018867924, 'recall': 0.2961783439490446, 'f1': 0.29430379746835444, 'number': 314}, 'P': {'precision': 0.19686800894854586, 'recall': 0.2430939226519337, 'f1': 0.21755253399258342, 'number': 362}, 'overall_precision': 0.23660130718954248, 'overall_recall': 0.2677514792899408, 'overall_f1': 0.2512144344205413, 'overall_accuracy': 0.6872164564042011}


		-------------RUN 4-----------
			------------EPOCH 1---------------
Loss:  tensor(662.9670, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1136.0884, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1669.7383, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1442.5315, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(769.1710, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1913.1208, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2946.9341, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2358.7556, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2101.4182, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1573.3049, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1301.1143, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2056.6628, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2004.6283, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1569.4271, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1316.3265, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2194.0796, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1213.0806, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1015.2957, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1371.1533, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1738.7281, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1799.0466, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2511.5403, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1885.4795, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1600.8960, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1451.3149, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2300.7656, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(939.7156, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1913.7089, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1408.3553, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1470.2507, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1607.1978, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2730.2246, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2273.1743, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2799.2061, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1775.4551, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1562.1783, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1692.9490, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2332.3301, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2373.6345, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2023.3716, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2126.2407, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1856.1128, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2880.0984, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1584.0261, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1423.7207, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1993.9204, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(888.7057, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1489.3657, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1148.8872, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3733.8608, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1785.6753, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2081.5757, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1123.4695, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1670.8048, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1293.5557, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1604.0889, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1763.7217, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.05069124423963134, 'recall': 0.04263565891472868, 'f1': 0.04631578947368422, 'number': 258}, 'P': {'precision': 0.13106796116504854, 'recall': 0.08571428571428572, 'f1': 0.1036468330134357, 'number': 315}, 'overall_precision': 0.08983451536643026, 'overall_recall': 0.06631762652705062, 'overall_f1': 0.07630522088353416, 'overall_accuracy': 0.5390455531453362}
			------------EPOCH 2---------------
Loss:  tensor(498.2805, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(795.9471, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1141.1101, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(950.8260, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(655.6947, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1464.4658, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2351.7520, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1934.0037, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1834.7322, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1290.1368, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(970.9877, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1496.8414, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1576.4678, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1267.1389, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1033.4436, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1727.9246, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1026.2103, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(788.6212, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1133.9364, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1467.6326, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1379.5599, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1865.2307, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1610.2328, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1360.2903, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1161.9490, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1898.8230, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(759.3673, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1554.5199, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1151.7517, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1273.9187, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1339.4321, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2448.8230, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1857.4651, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2216.0544, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1456.7108, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1263.9590, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1426.3152, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2038.7151, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2060.6057, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1716.5154, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1803.2330, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1582.3553, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2379.6602, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1194.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1102.6812, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1741.4896, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(750.8149, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1166.9434, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(888.3166, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3189.6333, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1513.8784, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1702.0969, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(930.5122, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1321.4237, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1168.5425, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1233.1733, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1583.0642, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.08469055374592833, 'recall': 0.10077519379844961, 'f1': 0.09203539823008848, 'number': 258}, 'P': {'precision': 0.05825242718446602, 'recall': 0.01904761904761905, 'f1': 0.02870813397129187, 'number': 315}, 'overall_precision': 0.07804878048780488, 'overall_recall': 0.055846422338568937, 'overall_f1': 0.06510681586978637, 'overall_accuracy': 0.5013015184381778}
			------------EPOCH 3---------------
Loss:  tensor(403.6972, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(725.7347, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(954.9949, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(786.6754, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(579.9286, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1318.7222, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2117.0190, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1702.2119, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1611.0505, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1127.4220, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(840.1672, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1225.2140, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1329.8386, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1096.9250, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(864.1809, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1515.7717, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(863.8187, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(609.1432, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(947.6519, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1216.0759, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1089.4651, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1418.7478, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1351.8395, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1164.8215, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(919.1535, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1549.1204, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(598.4705, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1247.0225, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(933.8994, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1139.7340, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1085.4489, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2146.2476, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1514.9076, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1785.6049, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(956.7825, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(789.7216, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(917.6581, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1611.0186, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1680.7350, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1448.3667, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1404.3748, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1374.1093, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1890.9033, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(914.6354, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(879.3525, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1496.6377, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(595.4817, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(927.4042, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(717.3478, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2703.3057, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1198.6663, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1407.0078, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(714.1887, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1038.2579, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1024.1737, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1005.6615, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1288.9226, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.21397379912663755, 'recall': 0.18992248062015504, 'f1': 0.20123203285420943, 'number': 258}, 'P': {'precision': 0.125, 'recall': 0.0380952380952381, 'f1': 0.05839416058394161, 'number': 315}, 'overall_precision': 0.18769230769230769, 'overall_recall': 0.10645724258289703, 'overall_f1': 0.13585746102449886, 'overall_accuracy': 0.5187997107736804}
			------------EPOCH 4---------------
Loss:  tensor(307.3273, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(646.1595, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(720.8954, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(592.9784, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(462.9594, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(938.3091, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1648.8341, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1442.0610, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1282.1661, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(986.1028, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(742.6364, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1071.3820, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1142.6348, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(947.7758, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(746.0747, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1368.0630, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(710.7980, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(456.5256, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(736.2630, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1026.7417, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(916.8552, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1157.3107, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1093.0486, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1061.2449, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(622.7397, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1168.7290, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(471.7332, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(961.0855, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(698.9214, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(960.8113, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(897.9481, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1821.3844, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1202.8976, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1491.0193, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(624.1862, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(449.9726, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(637.5156, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1410.3489, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1528.7119, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1278.7373, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1026.6329, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1212.9990, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1496.2698, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(701.2919, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(689.5435, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1246.3508, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(459.8901, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(711.4981, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(545.6216, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2331.5093, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(896.0916, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1036.7219, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(525.4946, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(759.2007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(883.3486, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(750.4808, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1059.8059, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.17801047120418848, 'recall': 0.13178294573643412, 'f1': 0.15144766146993321, 'number': 258}, 'P': {'precision': 0.15789473684210525, 'recall': 0.05714285714285714, 'f1': 0.0839160839160839, 'number': 315}, 'overall_precision': 0.17049180327868851, 'overall_recall': 0.09075043630017451, 'overall_f1': 0.11845102505694759, 'overall_accuracy': 0.5259580621836587}
			------------EPOCH 5---------------
Loss:  tensor(201.1599, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(532.0374, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(495.2047, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(403.6300, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(349.6887, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(702.6986, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1355.7673, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1271.5481, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1052.8555, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(834.5643, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(621.7585, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(900.2327, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(947.8623, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(857.1181, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(601.7159, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1193.9043, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(541.4423, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(338.4056, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(572.6605, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(886.0138, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(773.3291, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(986.2018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(831.5681, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(945.9598, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(450.2960, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(912.1552, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(386.2344, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(703.6985, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(506.0585, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(709.3560, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(740.1025, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1480.0022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(948.4402, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1215.6239, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(427.2747, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(279.1553, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(492.3220, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1152.1711, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1260.9736, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1095.9462, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(780.1260, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1075.9458, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1321.1077, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(611.5701, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(535.5071, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1048.9668, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(395.9653, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(629.0519, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(391.4214, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2037.7390, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(708.3107, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(793.2518, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(390.1049, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(582.9744, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(806.4506, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(621.0881, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(818.8617, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.12044374009508717, 'recall': 0.29457364341085274, 'f1': 0.17097862767154107, 'number': 258}, 'P': {'precision': 0.04950495049504951, 'recall': 0.015873015873015872, 'f1': 0.024038461538461536, 'number': 315}, 'overall_precision': 0.11065573770491803, 'overall_recall': 0.14136125654450263, 'overall_f1': 0.12413793103448277, 'overall_accuracy': 0.514172089660159}
			------------EPOCH 6---------------
Loss:  tensor(216.4716, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(422.5995, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(424.4094, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(339.7495, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(296.2695, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(957.0969, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1612.1786, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1463.4282, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(941.2028, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(626.9895, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(571.2729, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(823.3861, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1071.5712, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(876.4007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(685.6571, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1345.3180, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(512.0085, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(446.6019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(731.0067, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(920.7507, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(885.0566, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1311.0334, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(957.5309, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1036.6388, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(391.4450, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(907.0884, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(345.6707, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(638.4371, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(479.3265, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(600.0527, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(739.5496, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1395.0818, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(897.6158, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1253.3682, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(388.4991, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(230.8781, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(296.3515, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1002.5293, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1236.5449, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1055.7172, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(847.5158, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(924.2584, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1098.6660, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(645.3741, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(692.8324, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1246.3538, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(478.5979, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(812.5868, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(512.2219, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3407.5022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1172.8617, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1582.8774, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(642.0083, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1103.9043, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1002.4618, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1187.1482, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(906.9100, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.19230769230769232, 'recall': 0.2131782945736434, 'f1': 0.20220588235294115, 'number': 258}, 'P': {'precision': 0.13333333333333333, 'recall': 0.19682539682539682, 'f1': 0.15897435897435896, 'number': 315}, 'overall_precision': 0.15579227696404793, 'overall_recall': 0.20418848167539266, 'overall_f1': 0.17673716012084592, 'overall_accuracy': 0.5995661605206074}
			------------EPOCH 7---------------
Loss:  tensor(196.9632, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(325.6046, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(364.3727, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(334.6122, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(224.1176, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(816.7136, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1154.1407, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1359.8829, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1127.8174, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(900.8373, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(514.0215, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(753.1187, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(940.2831, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(823.1493, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(547.4662, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1062.6483, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(454.0801, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(346.8089, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(525.6083, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(731.9139, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(854.3134, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1144.5100, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(777.7935, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(879.8511, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(626.6266, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1347.4039, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(415.2195, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(880.9705, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(689.4926, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(993.0634, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1288.2688, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1839.8469, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1174.4288, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1571.8534, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(407.6177, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(298.8865, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(247.1101, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1027.9346, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1325.9935, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1272.9805, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(528.5815, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(879.4870, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(979.7002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(579.5118, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(424.9004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(878.1624, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(257.9811, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(489.4702, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(309.9240, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1696.7473, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(586.2985, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(663.9158, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(303.4194, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(503.2596, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(587.4482, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(529.7098, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(810.4454, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.25833333333333336, 'recall': 0.12015503875968993, 'f1': 0.16402116402116404, 'number': 258}, 'P': {'precision': 0.17537313432835822, 'recall': 0.2984126984126984, 'f1': 0.2209165687426557, 'number': 315}, 'overall_precision': 0.19054878048780488, 'overall_recall': 0.2181500872600349, 'overall_f1': 0.2034174125305126, 'overall_accuracy': 0.546999276934201}
			------------EPOCH 8---------------
Loss:  tensor(189.2593, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(411.4737, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(425.5276, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(411.8347, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(250.9963, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(930.0092, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1331.3064, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1821.5967, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1376.3975, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1078.9980, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(490.1899, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(983.1572, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(862.4554, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(697.9066, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(362.6012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1197.3794, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(430.1312, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(218.1233, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(479.1326, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(905.4299, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(578.3130, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(870.9752, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(589.3076, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(578.5220, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(320.5726, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(729.6633, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(356.0015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(543.8406, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(453.4454, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(531.4587, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(819.2924, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1368.8706, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(902.3571, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1377.1458, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(295.7237, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(157.8985, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(164.2004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(555.9338, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(980.4964, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1009.8594, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(411.1049, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(647.2936, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(839.1152, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(549.2338, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(393.9825, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(712.5651, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(194.7144, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(402.5344, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(312.7799, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1879.0211, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(566.5926, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(578.1546, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(275.9214, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(458.1782, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(431.7114, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(435.6726, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(603.6100, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.20610687022900764, 'recall': 0.10465116279069768, 'f1': 0.13881748071979436, 'number': 258}, 'P': {'precision': 0.13682092555331993, 'recall': 0.21587301587301588, 'f1': 0.16748768472906403, 'number': 315}, 'overall_precision': 0.15127388535031847, 'overall_recall': 0.16579406631762653, 'overall_f1': 0.1582014987510408, 'overall_accuracy': 0.6120028922631959}
			------------EPOCH 9---------------
Loss:  tensor(66.0163, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(355.7575, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(221.8733, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(205.7752, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(144.4379, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(475.8473, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(776.0709, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(866.5196, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(581.3521, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(490.0063, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(306.4552, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(485.1534, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(621.2949, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(452.7811, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(226.1705, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(874.9998, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(264.8805, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(147.1463, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(324.4209, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(492.9113, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(354.0115, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(502.8291, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(394.8408, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(389.5572, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(198.9715, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(514.5635, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(217.8383, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(356.1107, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(296.5090, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(368.5126, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(420.0319, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(896.8671, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(603.0377, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(808.1599, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(232.8962, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(114.5752, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(123.1365, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(438.7394, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(792.0490, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(833.5491, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(364.9681, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(496.3208, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(546.2024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(400.8256, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(293.2989, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(578.5676, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(136.7040, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(339.3461, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(166.5240, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1203.7096, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(410.3293, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(405.3098, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(157.5819, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(212.8994, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(299.0882, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(276.0277, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(357.7739, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.23711340206185566, 'recall': 0.08914728682170543, 'f1': 0.1295774647887324, 'number': 258}, 'P': {'precision': 0.14396887159533073, 'recall': 0.23492063492063492, 'f1': 0.17852834740651385, 'number': 315}, 'overall_precision': 0.15875613747954173, 'overall_recall': 0.16928446771378708, 'overall_f1': 0.16385135135135137, 'overall_accuracy': 0.603470715835141}
			------------EPOCH 10---------------
Loss:  tensor(55.9424, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(330.8210, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(169.9702, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(152.0450, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(138.6051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(425.2001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(646.7164, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(792.5264, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(693.4159, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(517.3077, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(350.6283, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(736.7704, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(776.9722, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(513.8699, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(284.8285, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1101.9192, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(280.5484, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(197.2677, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(467.2077, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(639.4996, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(313.8856, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(453.3339, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(409.7262, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(360.9229, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(126.8020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(431.6157, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(128.1309, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(263.6938, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(193.7140, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(266.2213, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(314.3446, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(683.2600, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(462.1451, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(626.8476, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(147.5015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(73.6252, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(91.1945, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(307.7212, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(629.7148, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(718.4508, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(476.9455, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(714.2025, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(654.3850, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(408.1913, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(510.1771, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(778.9072, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(250.1319, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(574.2222, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(196.5485, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1241.3070, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(340.4839, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(400.7698, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(176.1004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(206.7120, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(362.5863, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(256.8224, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(265.9146, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.2159090909090909, 'recall': 0.14728682170542637, 'f1': 0.17511520737327188, 'number': 258}, 'P': {'precision': 0.1446028513238289, 'recall': 0.2253968253968254, 'f1': 0.1761786600496278, 'number': 315}, 'overall_precision': 0.1634182908545727, 'overall_recall': 0.19022687609075042, 'overall_f1': 0.17580645161290323, 'overall_accuracy': 0.6104121475054229}
			------------EPOCH 11---------------
Loss:  tensor(39.3657, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(214.7825, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(123.8732, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(106.2611, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(64.5352, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(364.8051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(457.7946, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(570.0630, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(367.2298, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(305.8152, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(199.2455, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(284.1285, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(508.3176, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(363.1866, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(183.0948, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(728.3397, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(266.1060, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(168.1895, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(353.7695, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(423.9955, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(366.9249, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(469.8121, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(505.0957, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(401.2845, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(170.1940, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(490.1048, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(180.1323, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(465.8397, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(262.7799, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(390.8154, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(430.0759, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1087.6219, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(486.5342, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(538.3235, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(180.4329, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(91.6654, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(104.9961, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(346.4703, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(661.9398, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(584.8060, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(227.1847, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(283.7293, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(410.7708, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(342.9696, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(193.8045, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(320.4990, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(67.0658, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(187.2873, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(89.5978, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(923.4655, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(247.3402, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(251.3930, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(95.0034, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(175.2208, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(290.2908, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(243.1119, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(392.0746, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.13368983957219252, 'recall': 0.29069767441860467, 'f1': 0.18315018315018317, 'number': 258}, 'P': {'precision': 0.16666666666666666, 'recall': 0.11746031746031746, 'f1': 0.1378026070763501, 'number': 315}, 'overall_precision': 0.14303959131545338, 'overall_recall': 0.19546247818499127, 'overall_f1': 0.16519174041297935, 'overall_accuracy': 0.5767172812725958}
			------------EPOCH 12---------------
Loss:  tensor(76.6323, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(262.0130, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(202.8566, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(125.8176, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(158.2441, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(595.9033, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(852.6165, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(945.9193, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(655.7532, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(330.8231, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(278.8006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(443.0452, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(603.1705, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(443.3216, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(169.2923, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(689.0752, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(215.4147, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(125.0424, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(298.7722, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(350.3459, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(220.1468, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(366.0619, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(250.4586, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(312.8148, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(84.6819, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(340.3621, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(84.5266, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(189.3822, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(141.7184, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(187.6603, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(230.5499, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(511.5015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(290.3576, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(363.5840, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(118.5749, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(70.2668, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(105.1185, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(252.5095, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(514.3392, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(503.5270, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(222.0945, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(279.2742, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(414.2863, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(273.7747, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(274.4148, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(349.9797, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(118.8950, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(269.6415, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(142.5646, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(984.6566, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(323.5706, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(291.3057, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(144.5118, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(175.5751, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(164.5496, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(250.5750, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(235.2645, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.21764705882352942, 'recall': 0.1434108527131783, 'f1': 0.17289719626168223, 'number': 258}, 'P': {'precision': 0.14285714285714285, 'recall': 0.2253968253968254, 'f1': 0.17487684729064037, 'number': 315}, 'overall_precision': 0.1619190404797601, 'overall_recall': 0.18848167539267016, 'overall_f1': 0.17419354838709677, 'overall_accuracy': 0.6130151843817787}
			------------EPOCH 13---------------
Loss:  tensor(26.9145, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(184.0737, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(110.9665, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(85.9665, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.7954, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(329.3596, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(360.1268, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(397.0956, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(237.5100, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(176.2845, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(109.6428, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(186.2150, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(397.4957, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(258.5933, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(74.5431, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(463.3087, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(124.8681, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.0031, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(137.9783, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(159.1149, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(136.7554, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(237.7773, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(173.3394, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(194.9500, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(66.5971, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(261.2126, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(93.3154, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(174.5894, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(112.1800, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(169.8822, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(187.7379, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(478.6110, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(269.5776, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(359.8914, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(78.6368, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.5511, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.0644, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(128.3189, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(399.5363, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(406.0568, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(148.9068, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(135.5660, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(259.6032, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(209.3653, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(101.4772, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(189.1912, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.0140, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(136.6578, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.6225, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(684.0804, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(227.7934, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(192.3440, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.5032, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(88.1689, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(115.4983, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(130.5990, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(143.2885, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.1875, 'recall': 0.18604651162790697, 'f1': 0.1867704280155642, 'number': 258}, 'P': {'precision': 0.13260869565217392, 'recall': 0.19365079365079366, 'f1': 0.15741935483870972, 'number': 315}, 'overall_precision': 0.15223463687150837, 'overall_recall': 0.19022687609075042, 'overall_f1': 0.1691233514352211, 'overall_accuracy': 0.6184381778741865}
			------------EPOCH 14---------------
Loss:  tensor(14.4051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(149.1795, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.4026, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.2882, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.6342, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(259.6963, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(347.8601, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(363.4194, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(160.6358, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(129.3727, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(86.8640, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(125.0193, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(320.5979, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(153.9969, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.9454, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(378.1823, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(94.4154, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.9627, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(100.2253, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(97.7455, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(98.2736, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(185.6537, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(142.7788, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(115.9007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.6570, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(206.3151, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.8851, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(131.6089, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(88.3153, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(127.7921, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(149.1370, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(353.3759, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(211.3473, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(298.8508, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(64.8322, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.9442, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.8415, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(92.0320, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(327.2798, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(345.6131, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(88.9616, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(77.4340, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(191.5398, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(173.0627, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(70.0664, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(129.8715, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.0329, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(107.2772, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.7458, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(593.4482, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(171.5169, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(151.8400, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.9520, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(59.5733, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(79.1134, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(88.4857, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(130.5052, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.19421487603305784, 'recall': 0.1821705426356589, 'f1': 0.18799999999999997, 'number': 258}, 'P': {'precision': 0.13260869565217392, 'recall': 0.19365079365079366, 'f1': 0.15741935483870972, 'number': 315}, 'overall_precision': 0.15384615384615385, 'overall_recall': 0.18848167539267016, 'overall_f1': 0.16941176470588237, 'overall_accuracy': 0.6080983369486623}
			------------EPOCH 15---------------
Loss:  tensor(13.7627, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(118.6975, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.1023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.3037, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.2788, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(246.4023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(283.9539, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(307.6340, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(133.7655, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(92.3751, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(64.9538, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(98.5579, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(270.6234, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(122.3668, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.6967, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(315.2374, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(67.1751, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.2245, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(77.4377, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(71.9320, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(74.5517, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(136.8178, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(84.5964, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(115.7472, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.4769, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(162.3728, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.5038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(94.7424, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(61.1781, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(102.2880, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(125.7710, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(304.2294, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(208.9154, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(321.2966, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.6263, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.4984, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.2330, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(78.0091, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(302.6726, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(303.7274, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.8245, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(69.4751, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(171.7553, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(140.5543, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.9864, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(111.9251, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.5476, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(79.0009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.8849, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(498.1375, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(145.8700, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(125.4755, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.1305, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.2637, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(59.2888, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(70.9410, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(100.7010, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.19747899159663865, 'recall': 0.1821705426356589, 'f1': 0.1895161290322581, 'number': 258}, 'P': {'precision': 0.15301724137931033, 'recall': 0.2253968253968254, 'f1': 0.18228498074454427, 'number': 315}, 'overall_precision': 0.16809116809116809, 'overall_recall': 0.20593368237347295, 'overall_f1': 0.18509803921568627, 'overall_accuracy': 0.6148951554591467}
			------------EPOCH 16---------------
Loss:  tensor(9.1062, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(90.9470, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.3122, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.7480, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.9919, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(219.4763, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(254.0502, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(280.7676, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(94.7509, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.5028, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.8018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(81.6059, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(239.1132, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(99.8010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.9307, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(264.2442, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.4958, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.1299, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.4367, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.3930, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.5044, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(106.8255, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(86.8788, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(68.9345, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.2356, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(143.5943, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.0450, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(85.2410, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.5506, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(90.1618, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(97.6409, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(262.4336, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(149.2826, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(218.8855, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.6348, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.1463, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.8730, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.7603, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(271.4467, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(263.1403, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.3413, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.6496, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(152.1831, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(120.3149, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.1160, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(86.3982, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.1246, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(64.8546, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.4183, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(466.7064, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(143.0927, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(108.2774, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.7220, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.6338, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.8274, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.7470, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(83.4853, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.20784313725490197, 'recall': 0.2054263565891473, 'f1': 0.20662768031189085, 'number': 258}, 'P': {'precision': 0.14316239316239315, 'recall': 0.2126984126984127, 'f1': 0.1711366538952746, 'number': 315}, 'overall_precision': 0.16597510373443983, 'overall_recall': 0.2094240837696335, 'overall_f1': 0.18518518518518517, 'overall_accuracy': 0.6139551699204627}
			------------EPOCH 17---------------
Loss:  tensor(6.5716, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(67.0384, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.3333, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.3451, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.6695, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(202.7117, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(240.0030, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(261.6187, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(64.4224, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.4304, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.0006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.3406, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(206.6880, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(78.0137, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.3832, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(216.1696, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.3344, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.6447, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.1946, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.1064, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.6887, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(78.3601, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.5405, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(77.7340, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.2775, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(125.3331, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.0744, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.2533, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.1084, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(70.4682, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(86.6905, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(233.3152, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(121.5583, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(194.8194, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.3534, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.1663, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.7684, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.1085, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(245.1119, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(235.9596, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.9259, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.3359, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(152.7185, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(107.5552, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.7682, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(69.7042, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.5661, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.1146, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.1428, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(424.5851, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(126.7875, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(94.0726, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.4435, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.4248, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.0327, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.1093, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(72.7144, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.23770491803278687, 'recall': 0.2248062015503876, 'f1': 0.23107569721115537, 'number': 258}, 'P': {'precision': 0.1328976034858388, 'recall': 0.19365079365079366, 'f1': 0.15762273901808788, 'number': 315}, 'overall_precision': 0.16927453769559034, 'overall_recall': 0.20767888307155322, 'overall_f1': 0.1865203761755486, 'overall_accuracy': 0.6141720896601591}
			------------EPOCH 18---------------
Loss:  tensor(4.7906, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.8916, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.0255, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.8809, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.7228, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(183.8914, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(219.9790, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(249.0867, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.3672, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.8644, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.6172, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.7956, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(188.0406, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.1072, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.4745, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(191.2270, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.9986, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.4462, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.5121, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.9385, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.3165, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(65.4466, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.3775, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.0743, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.2054, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(124.0017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.3559, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.8540, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.8670, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(65.7534, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(88.0416, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(217.7061, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(96.7656, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(226.4499, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.5318, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.3136, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.2061, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.0518, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(246.8550, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(234.8979, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.3052, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.6332, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(149.8705, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(102.9769, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.1619, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.6407, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.9995, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.0329, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.8175, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(399.2671, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(117.1466, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(86.5825, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.3440, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.3906, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.3356, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.2891, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(70.0855, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.24561403508771928, 'recall': 0.21705426356589147, 'f1': 0.23045267489711935, 'number': 258}, 'P': {'precision': 0.12008281573498965, 'recall': 0.18412698412698414, 'f1': 0.14536340852130328, 'number': 315}, 'overall_precision': 0.16033755274261605, 'overall_recall': 0.19895287958115182, 'overall_f1': 0.17757009345794392, 'overall_accuracy': 0.6075921908893709}
			------------EPOCH 19---------------
Loss:  tensor(3.6622, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.5941, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.3959, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.9476, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.5639, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(297.9063, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(283.7346, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(280.4700, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.9437, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.9015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.0755, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.5489, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(181.4934, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.9473, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.4194, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(193.1001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.5128, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.1069, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.5951, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.7846, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.8852, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.1322, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.9038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.9682, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.1318, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(111.4749, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.4443, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.4676, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.4205, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(66.7116, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(75.7954, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(211.4448, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(87.1956, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(209.7578, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.6605, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.3368, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.7944, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.0686, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(256.9941, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(231.1794, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.1956, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.3682, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(128.3600, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(108.5100, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.1343, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(67.2125, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.3940, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.9618, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.3938, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(434.8405, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(113.5004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(82.5163, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.3162, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.7069, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.1597, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.3515, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(79.6144, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.20634920634920634, 'recall': 0.20155038759689922, 'f1': 0.203921568627451, 'number': 258}, 'P': {'precision': 0.11373390557939914, 'recall': 0.16825396825396827, 'f1': 0.1357234314980794, 'number': 315}, 'overall_precision': 0.14623955431754876, 'overall_recall': 0.18324607329842932, 'overall_f1': 0.16266460108443068, 'overall_accuracy': 0.6141720896601591}
			------------EPOCH 20---------------
Loss:  tensor(4.5259, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.3050, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.3785, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.8064, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.5430, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(256.3133, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(269.3454, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(256.1593, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.3040, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.4234, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.1177, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.1021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(167.9499, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.2222, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.7241, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(163.1437, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.9006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.0526, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.2638, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.5884, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.2210, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.0072, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.0191, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.7236, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.7345, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(95.2152, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.7138, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.2534, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.6631, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.8797, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(69.4924, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(192.6894, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(75.4199, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(159.4157, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.6988, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.1437, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.6609, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.7499, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(228.8171, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(214.8645, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.3726, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.3868, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(121.6868, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(100.2122, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.4122, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.3062, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.0188, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.4493, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.6801, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(365.1325, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(98.1128, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(70.5815, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.4188, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.4462, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.0909, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.9489, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.7649, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.2222222222222222, 'recall': 0.20930232558139536, 'f1': 0.21556886227544908, 'number': 258}, 'P': {'precision': 0.13062098501070663, 'recall': 0.19365079365079366, 'f1': 0.15601023017902813, 'number': 315}, 'overall_precision': 0.1619718309859155, 'overall_recall': 0.2006980802792321, 'overall_f1': 0.17926734216679657, 'overall_accuracy': 0.6120751988430947}
			------------EPOCH 21---------------
Loss:  tensor(2.1551, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.7272, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.2698, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.2222, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.7908, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(194.6884, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(223.0241, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(234.0687, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.8124, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.8507, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.2682, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.6681, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(144.1126, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.3929, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.9786, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(134.2642, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.4511, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.5375, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.0409, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.1612, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.1244, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.3318, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.6293, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.5423, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.0191, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(77.7822, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.1507, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.7623, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.7261, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.1360, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(61.4064, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(172.1288, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(62.5605, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(134.2845, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.3684, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.9854, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.7195, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.4233, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(198.8177, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(189.0099, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.5619, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.9891, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(110.6027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(77.7665, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.8030, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.4760, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.4518, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.8966, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.5205, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(332.5153, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(91.9199, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(62.4519, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.8396, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.9199, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.5181, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.7442, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.4595, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.22448979591836735, 'recall': 0.2131782945736434, 'f1': 0.2186878727634195, 'number': 258}, 'P': {'precision': 0.12266112266112267, 'recall': 0.1873015873015873, 'f1': 0.14824120603015076, 'number': 315}, 'overall_precision': 0.15702479338842976, 'overall_recall': 0.19895287958115182, 'overall_f1': 0.17551963048498845, 'overall_accuracy': 0.6155459146782357}
			------------EPOCH 22---------------
Loss:  tensor(1.9402, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.8762, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.6978, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.0043, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.6220, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(162.8306, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(195.7527, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(222.2787, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.5475, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.6666, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.3389, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.1748, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(141.5159, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.6078, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.0465, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(132.6079, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.3738, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.7927, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.7906, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.8730, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.0645, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.4251, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(61.4765, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.0048, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.1784, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(75.1595, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.1929, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.3686, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.7867, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.8324, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(62.1119, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(161.3666, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.9105, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(125.2925, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.3823, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.5241, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.9516, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.5793, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(174.9953, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(169.1555, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.5920, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.2677, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(105.9360, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(72.4080, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.0371, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.2115, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.1430, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.7386, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.1657, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(316.9142, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(98.8871, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(64.4773, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.3005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.5845, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.2839, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.6848, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.5872, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.21484375, 'recall': 0.2131782945736434, 'f1': 0.2140077821011673, 'number': 258}, 'P': {'precision': 0.13043478260869565, 'recall': 0.19047619047619047, 'f1': 0.15483870967741933, 'number': 315}, 'overall_precision': 0.16061452513966482, 'overall_recall': 0.2006980802792321, 'overall_f1': 0.17843289371605897, 'overall_accuracy': 0.6096167751265366}
			------------EPOCH 23---------------
Loss:  tensor(1.2203, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.7663, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.8168, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.3575, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.0980, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(161.2690, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(205.3626, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(239.0474, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.5749, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.6838, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.1386, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.8286, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(134.2652, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.2114, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.0379, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(139.0768, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.0932, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.2029, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.1902, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.6079, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.2173, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(62.3743, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.3969, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(102.5564, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.2693, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(77.2776, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.0447, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.6192, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.0615, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.2724, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(70.5229, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(153.9960, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(61.0856, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(136.6731, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.1104, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.7874, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.2014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.6846, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(174.3924, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(161.0430, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.4153, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.2737, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(119.8173, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(69.3295, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.8787, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.2979, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.0771, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.1513, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.4751, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(335.4356, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(100.9280, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(65.8422, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.6644, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.4206, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.7930, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.9761, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.4059, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.2076271186440678, 'recall': 0.18992248062015504, 'f1': 0.19838056680161945, 'number': 258}, 'P': {'precision': 0.10826771653543307, 'recall': 0.1746031746031746, 'f1': 0.13365735115431351, 'number': 315}, 'overall_precision': 0.13978494623655913, 'overall_recall': 0.18150087260034903, 'overall_f1': 0.15793470007593013, 'overall_accuracy': 0.6145336225596529}
			------------EPOCH 24---------------
Loss:  tensor(1.1640, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.0215, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.3172, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.9812, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.4537, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(168.9032, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(210.7639, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(228.3434, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.0608, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.5920, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.6719, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.4956, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(143.6258, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.1916, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.0062, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(125.9234, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.8551, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.3352, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.9325, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.8532, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.1873, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.9581, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.7240, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.2041, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.3538, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(71.7317, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.8672, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.9322, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.9483, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.5239, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(59.0794, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(153.8368, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(123.3181, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(260.5380, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.6103, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.1415, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.0525, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.6647, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(167.5372, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(166.1085, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.6768, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.5817, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(101.2351, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(72.7307, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.7039, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.7833, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.6592, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.5143, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.5904, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(347.6470, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(89.5763, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(68.0819, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.9270, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.7837, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.4842, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.9759, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.7558, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.210727969348659, 'recall': 0.2131782945736434, 'f1': 0.2119460500963391, 'number': 258}, 'P': {'precision': 0.11740890688259109, 'recall': 0.18412698412698414, 'f1': 0.1433868974042027, 'number': 315}, 'overall_precision': 0.14966887417218544, 'overall_recall': 0.19720767888307156, 'overall_f1': 0.17018072289156627, 'overall_accuracy': 0.6040491684743312}
			------------EPOCH 25---------------
Loss:  tensor(1.1520, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.3453, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.6883, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.4606, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.8253, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(154.8835, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(190.3306, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(208.0604, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.0102, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.2961, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.1667, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.3751, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(123.0550, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.3589, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.0591, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(122.1046, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.5942, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.9396, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.1018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.3638, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.2557, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.4831, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.6001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.6919, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.9487, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(62.5163, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.3637, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.8243, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.5954, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.1469, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.1679, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(170.2001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(82.0591, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(184.3634, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.7233, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.2852, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.1233, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.8383, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(169.6451, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(157.3394, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.5559, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.7920, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(107.4238, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(69.7260, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.3794, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.7962, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.9557, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.9357, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.3076, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(363.2110, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(124.2678, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(83.7046, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.1384, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.8860, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.5383, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.2186, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.9799, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.21374045801526717, 'recall': 0.21705426356589147, 'f1': 0.2153846153846154, 'number': 258}, 'P': {'precision': 0.12727272727272726, 'recall': 0.17777777777777778, 'f1': 0.14834437086092717, 'number': 315}, 'overall_precision': 0.15954415954415954, 'overall_recall': 0.19546247818499127, 'overall_f1': 0.17568627450980392, 'overall_accuracy': 0.6077368040491685}
			------------EPOCH 26---------------
Loss:  tensor(2.3185, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.8561, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.1458, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.1405, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.8846, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(158.4158, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(190.8582, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(214.3548, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.6290, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.1197, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.3087, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.4771, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(125.1101, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.0186, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.3751, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(117.7219, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.4577, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.4680, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.9899, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.4428, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.3304, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.1860, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(59.0666, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.4004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.1349, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(80.4863, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.9494, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.7227, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.7009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.2033, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(65.9948, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(203.5394, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(64.4177, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(173.4499, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.1229, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.8845, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.0005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.9859, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(204.1994, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(202.1945, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.8948, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.4854, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(105.8768, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(79.0034, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.4614, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.6270, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.2812, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.3135, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.5331, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(350.5535, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(220.7892, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(119.7900, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.0729, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.0194, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.7944, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.2302, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.3831, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.22748815165876776, 'recall': 0.18604651162790697, 'f1': 0.20469083155650317, 'number': 258}, 'P': {'precision': 0.1232876712328767, 'recall': 0.2, 'f1': 0.15254237288135591, 'number': 315}, 'overall_precision': 0.15373961218836565, 'overall_recall': 0.193717277486911, 'overall_f1': 0.17142857142857143, 'overall_accuracy': 0.590527838033261}
			------------EPOCH 27---------------
Loss:  tensor(1.0257, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.0550, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.1085, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.2021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.5159, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(214.6737, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(243.0353, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(258.0355, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.2768, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.6887, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.0224, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.7865, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(178.9279, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.6216, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.7621, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(166.8705, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.6907, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.5860, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.9150, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.9606, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.6310, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.9294, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(64.7207, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.6857, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.8555, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(71.2172, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.4739, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.8250, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.1864, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.5242, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.2450, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(177.9921, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(73.1279, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(179.5991, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.2099, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.9361, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.1752, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.8646, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(224.1154, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(193.4930, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.6882, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.3039, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(114.5133, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(83.3491, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.1128, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.5237, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.9060, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.6213, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.6178, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(398.0059, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(162.4443, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(98.3133, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.0607, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.2731, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.1347, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.0029, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.3827, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.24358974358974358, 'recall': 0.22093023255813954, 'f1': 0.23170731707317077, 'number': 258}, 'P': {'precision': 0.12159329140461216, 'recall': 0.18412698412698414, 'f1': 0.14646464646464646, 'number': 315}, 'overall_precision': 0.1617440225035162, 'overall_recall': 0.2006980802792321, 'overall_f1': 0.17912772585669784, 'overall_accuracy': 0.6046999276934201}
			------------EPOCH 28---------------
Loss:  tensor(0.7264, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.4031, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.8526, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.8987, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.0454, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(188.2655, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(205.5231, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(212.7307, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.7261, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.3229, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.0485, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.0683, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(125.5195, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.6158, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.1500, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(108.3953, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.4547, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.4960, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.8067, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.3999, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.8471, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.4685, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.4020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(61.8209, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.7474, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(88.4129, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(70.2450, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(163.5040, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.5563, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.1159, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.4233, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(162.7993, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.8139, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(115.0665, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.2972, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.5092, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.5564, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.8098, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(168.8765, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(149.4170, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.5965, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.0447, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(92.1823, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(74.5568, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.4831, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.7795, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.6327, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.5125, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.9657, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(307.2213, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(106.2846, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(64.9000, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.0252, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.8958, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.0247, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.7237, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.2827, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.19140625, 'recall': 0.18992248062015504, 'f1': 0.19066147859922178, 'number': 258}, 'P': {'precision': 0.13806706114398423, 'recall': 0.2222222222222222, 'f1': 0.170316301703163, 'number': 315}, 'overall_precision': 0.1559633027522936, 'overall_recall': 0.20767888307155322, 'overall_f1': 0.17814371257485032, 'overall_accuracy': 0.6153289949385394}
			------------EPOCH 29---------------
Loss:  tensor(8.8927, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.5818, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.9338, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.5033, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.2055, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(163.8348, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(190.8459, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(194.6513, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.4896, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.8552, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.6807, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.2525, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(115.0432, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.0082, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.0629, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(97.7810, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.1173, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.6057, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.6281, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.0999, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.6160, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.5019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.7942, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.6281, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.0292, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(69.9040, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.4544, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.9727, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.6082, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.1193, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.0545, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(151.1084, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.5848, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(134.0451, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.3155, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.2927, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.8441, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(73.2130, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(174.1266, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(140.7630, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.3204, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.3423, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(87.6938, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(65.4069, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.1199, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.9037, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.8207, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.3886, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.4815, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(307.1194, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(93.4437, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(65.6647, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.6953, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.6535, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.8400, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.6229, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.6810, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.2054263565891473, 'recall': 0.2054263565891473, 'f1': 0.2054263565891473, 'number': 258}, 'P': {'precision': 0.11513859275053305, 'recall': 0.17142857142857143, 'f1': 0.1377551020408163, 'number': 315}, 'overall_precision': 0.14718019257221457, 'overall_recall': 0.18673647469458987, 'overall_f1': 0.1646153846153846, 'overall_accuracy': 0.6022415039768619}
			------------EPOCH 30---------------
Loss:  tensor(0.9873, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.0941, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.7944, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.3754, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.8883, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(125.7716, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(161.8312, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(179.5550, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.0175, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.3848, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.3537, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.9956, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(101.5087, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.4739, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.8581, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(87.7702, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.0312, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.6644, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.5646, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.3608, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.3895, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.3120, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.0962, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.2376, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.7462, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(65.1192, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.8048, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.9255, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.2227, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.1772, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.9399, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(133.3797, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.6401, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(108.4816, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.1416, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.8458, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5290, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(71.9504, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(159.6365, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(111.7177, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.2540, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.4203, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(83.3797, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.7465, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.8900, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.0220, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.3644, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.6349, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.6561, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(236.9098, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(72.0220, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.0561, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.1016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.6886, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.6142, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.2440, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.9033, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.18840579710144928, 'recall': 0.20155038759689922, 'f1': 0.1947565543071161, 'number': 258}, 'P': {'precision': 0.10412573673870335, 'recall': 0.16825396825396827, 'f1': 0.12864077669902912, 'number': 315}, 'overall_precision': 0.1337579617834395, 'overall_recall': 0.18324607329842932, 'overall_f1': 0.15463917525773196, 'overall_accuracy': 0.5975415762834418}


		-------------RUN 5-----------
			------------EPOCH 1---------------
Loss:  tensor(3327.6541, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1828.7256, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1593.1085, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1430.9958, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(953.7803, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1489.6293, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2144.6831, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1256.2976, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3115.2686, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2281.1404, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1755.0070, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1562.9785, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1966.7629, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1754.5342, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2049.3047, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1726.3278, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1202.0533, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(935.6795, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1800.7346, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2073.0115, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1227.9207, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2448.1641, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2062.3069, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1067.7434, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1621.9346, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1471.8347, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1322.6399, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1232.4861, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1865.2532, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1792.7520, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1848.8403, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1026.4910, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1307.9951, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1310.4961, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1237.8315, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2164.8691, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1022.5795, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1603.5603, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1795.1749, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1414.9698, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1755.5543, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2371.9150, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1572.2727, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1837.7252, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1323.0767, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1864.5065, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1512.2000, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2279.6187, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2048.3323, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2362.7273, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1131.8533, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1005.2738, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1237.5389, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1702.5876, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1711.4658, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1344.6346, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1677.2400, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'number': 268}, 'P': {'precision': 0.09032258064516129, 'recall': 0.16568047337278108, 'f1': 0.11691022964509394, 'number': 338}, 'overall_precision': 0.08602150537634409, 'overall_recall': 0.0924092409240924, 'overall_f1': 0.0891010342084328, 'overall_accuracy': 0.5096371569006461}
			------------EPOCH 2---------------
Loss:  tensor(2037.4939, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1069.4418, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1011.5187, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(859.1134, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(698.7287, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(902.8551, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1549.0598, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(943.5805, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2531.1611, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1754.3259, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1329.6736, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1163.7356, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1594.0278, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1405.7183, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1547.0582, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1423.9556, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(970.9154, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(733.9524, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1520.7379, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1724.9800, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1028.6340, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2035.5793, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1913.2644, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(853.4561, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1349.5645, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1233.4020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1151.0554, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1087.7715, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1412.9373, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1417.8718, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1544.1484, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(866.1786, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1073.6801, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1107.8345, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1045.7092, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1761.6255, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(866.4966, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1339.8345, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1484.4711, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1227.4568, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1367.3119, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1987.0725, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1336.8995, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1558.1545, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1134.5740, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1609.1067, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1273.9758, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2008.7853, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1734.3705, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2053.3279, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1043.7312, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(841.7582, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1047.2217, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1527.2129, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1482.7783, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1147.7850, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1534.0496, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.041666666666666664, 'recall': 0.022388059701492536, 'f1': 0.029126213592233007, 'number': 268}, 'P': {'precision': 0.06439393939393939, 'recall': 0.10059171597633136, 'f1': 0.07852193995381061, 'number': 338}, 'overall_precision': 0.05952380952380952, 'overall_recall': 0.066006600660066, 'overall_f1': 0.06259780907668232, 'overall_accuracy': 0.5484619208041089}
			------------EPOCH 3---------------
Loss:  tensor(1768.8994, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(887.8158, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(820.9521, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(691.2745, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(618.5703, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(712.8729, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1250.0549, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(783.6932, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2244.9453, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1449.1643, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1067.2836, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(923.7551, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1472.7087, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1186.6205, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1209.7360, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1207.1320, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(805.3080, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(592.7964, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1337.8712, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1461.5977, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(868.9504, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1718.3136, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1627.4156, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(705.1401, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1082.5103, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1048.9833, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(989.2996, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(917.9886, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1218.5322, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1241.0133, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1432.4872, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(707.7474, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(896.1718, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(927.2861, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(887.5739, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1324.8557, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(638.2460, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(908.2656, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(974.5701, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1063.3984, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(964.6073, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1578.7098, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1109.1252, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1389.0546, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(920.8988, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1319.9969, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1082.4309, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1697.3013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1501.7754, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1792.6360, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(939.2852, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(704.4944, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(934.5681, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1360.6846, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1307.8096, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(978.0378, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1329.5156, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.09282700421940929, 'recall': 0.08208955223880597, 'f1': 0.08712871287128714, 'number': 268}, 'P': {'precision': 0.11970074812967581, 'recall': 0.14201183431952663, 'f1': 0.12990527740189445, 'number': 338}, 'overall_precision': 0.109717868338558, 'overall_recall': 0.11551155115511551, 'overall_f1': 0.11254019292604502, 'overall_accuracy': 0.5836416855359806}
			------------EPOCH 4---------------
Loss:  tensor(1651.2734, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(824.5779, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(730.1278, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(609.6375, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(509.8575, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(598.0051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1103.5054, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(682.2770, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1913.3883, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1278.9448, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(935.4879, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(821.6085, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1238.4858, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(947.0066, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(965.8030, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(897.4261, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(649.0684, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(504.0909, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1124.4374, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1238.3397, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(772.6073, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1572.9924, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1418.7159, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(593.6896, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(899.4484, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(907.6789, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(905.9660, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(832.9377, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1110.1896, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1173.4127, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1388.4509, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(594.0355, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(692.8798, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(817.2621, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(786.8180, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1034.0205, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(446.2600, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(668.1844, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(674.8571, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(803.8369, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(764.1738, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1299.3868, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(899.3972, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1146.8419, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(764.1617, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1154.9038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(918.4232, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1365.3457, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1250.0470, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1622.7910, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(798.1217, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(587.1061, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(812.5238, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1173.3091, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1187.0663, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(805.4802, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1234.2212, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.10541310541310542, 'recall': 0.13805970149253732, 'f1': 0.11954765751211632, 'number': 268}, 'P': {'precision': 0.10555555555555556, 'recall': 0.11242603550295859, 'f1': 0.10888252148997135, 'number': 338}, 'overall_precision': 0.10548523206751055, 'overall_recall': 0.12376237623762376, 'overall_f1': 0.11389521640091116, 'overall_accuracy': 0.5559728281879935}
			------------EPOCH 5---------------
Loss:  tensor(1430.3085, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(695.2889, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(558.9109, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(435.2716, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(445.9667, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(448.6496, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(930.7693, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(603.5438, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1698.4968, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1055.9224, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(717.9374, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(657.1572, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1095.2877, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(801.1673, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(798.8567, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(723.5159, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(520.0038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(378.2147, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(986.6807, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1121.3014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(658.4089, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1299.1311, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1071.8654, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(417.2009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(608.9473, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(690.6186, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(757.4178, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(665.8755, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(994.7178, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1137.7310, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1336.2412, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(532.5125, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(585.5605, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(762.5365, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(726.5319, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(921.4670, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(405.7488, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(595.8365, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(611.6632, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(895.3481, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(653.7427, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(996.0568, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(652.6295, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1066.3982, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(623.3348, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(997.2399, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(755.6446, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1051.9631, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1057.0635, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1328.9299, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(574.1981, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(497.0896, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(716.6603, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(929.5645, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(981.4938, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(746.5112, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(973.2372, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.14555765595463138, 'recall': 0.2873134328358209, 'f1': 0.19322459222082808, 'number': 268}, 'P': {'precision': 0.13761467889908258, 'recall': 0.04437869822485207, 'f1': 0.06711409395973154, 'number': 338}, 'overall_precision': 0.14420062695924765, 'overall_recall': 0.15181518151815182, 'overall_f1': 0.14790996784565913, 'overall_accuracy': 0.4875462528303971}
			------------EPOCH 6---------------
Loss:  tensor(1442.7952, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(742.8188, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(543.2167, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(421.6436, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(471.8313, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(468.3620, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(934.1091, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(608.5137, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1425.9377, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(946.5141, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(626.6329, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(593.9511, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(901.8433, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(765.2167, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(643.9501, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(569.1062, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(423.3263, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(257.7215, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(665.1352, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(842.9939, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(556.1821, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1049.5406, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(910.0886, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(310.7071, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(428.1940, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(544.3710, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(580.1057, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(526.9451, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(869.2920, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1021.7762, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1308.8070, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(466.6440, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(486.3923, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(726.5508, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(728.1553, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(706.4012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(296.0790, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(513.0158, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(543.1768, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(820.6527, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(572.0098, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(867.6759, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(553.9194, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(968.9216, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(622.4615, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(995.2228, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(727.5715, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(904.9828, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(819.9517, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1139.8845, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(520.5671, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(417.2126, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(547.0148, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(759.2943, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(929.0574, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(558.8937, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(832.9550, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.1398176291793313, 'recall': 0.34328358208955223, 'f1': 0.19870410367170627, 'number': 268}, 'P': {'precision': 0.21212121212121213, 'recall': 0.04142011834319527, 'f1': 0.06930693069306931, 'number': 338}, 'overall_precision': 0.1464088397790055, 'overall_recall': 0.17491749174917492, 'overall_f1': 0.15939849624060148, 'overall_accuracy': 0.4619208041089082}
			------------EPOCH 7---------------
Loss:  tensor(1490.7504, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(739.8674, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(426.8768, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(279.6396, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(514.9343, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(695.3954, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1257.2260, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(642.2019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1636.0598, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1063.6737, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(703.0909, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(712.3329, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(967.7570, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(783.6876, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(844.1058, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(704.9412, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(344.8151, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(221.4549, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(650.7531, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1015.2605, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(559.0533, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(966.2922, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(714.7148, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(285.5469, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(359.4888, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(423.8101, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(502.5018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(421.4183, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(759.0460, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(894.5419, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1157.2607, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(404.8529, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(451.8936, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(668.1373, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(685.5336, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(662.0269, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(292.4495, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(542.1105, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(573.9929, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(776.5830, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(580.5281, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(979.3322, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(653.4406, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1319.3303, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(824.0497, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1277.1317, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1110.8419, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1348.2118, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(950.7349, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1405.0446, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(700.3504, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(523.8904, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(542.2878, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(913.2767, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1173.6534, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(581.3253, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(952.5569, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.18271604938271604, 'recall': 0.27611940298507465, 'f1': 0.21991084695393762, 'number': 268}, 'P': {'precision': 0.12010443864229765, 'recall': 0.13609467455621302, 'f1': 0.1276005547850208, 'number': 338}, 'overall_precision': 0.15228426395939088, 'overall_recall': 0.19801980198019803, 'overall_f1': 0.17216642754662842, 'overall_accuracy': 0.5334401060363395}
			------------EPOCH 8---------------
Loss:  tensor(1184.0719, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(433.7566, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(321.6756, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(214.5095, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(419.7172, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(478.8915, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(775.4399, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(503.4858, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1225.6373, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(815.9724, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(524.9778, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(486.2540, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(866.5660, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(669.1548, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(752.4691, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(594.5707, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(424.6555, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(243.9701, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(717.8370, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1129.5729, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(601.3427, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1175.9399, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1060.5050, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(298.4823, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(490.5426, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(594.5477, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(620.0290, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(552.0322, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(942.1308, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(771.9093, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(990.8516, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(357.0370, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(417.1601, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(479.4917, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(420.1334, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(671.4004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(225.4402, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(308.9758, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(245.6713, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(480.3423, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(510.3984, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(721.9052, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(428.8808, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(612.3358, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(449.9883, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(687.9105, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(622.3143, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(714.3008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(657.4062, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(866.6296, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(435.9977, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(433.3806, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(475.0450, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(664.2018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(898.6204, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(592.7516, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(985.2692, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.27976190476190477, 'recall': 0.17537313432835822, 'f1': 0.21559633027522934, 'number': 268}, 'P': {'precision': 0.2909090909090909, 'recall': 0.47337278106508873, 'f1': 0.36036036036036034, 'number': 338}, 'overall_precision': 0.2883008356545961, 'overall_recall': 0.3415841584158416, 'overall_f1': 0.31268882175226587, 'overall_accuracy': 0.5571878279118573}
			------------EPOCH 9---------------
Loss:  tensor(1071.5902, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(371.4100, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(351.1310, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(262.3472, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(310.7724, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(193.0336, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(598.1335, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(469.1861, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1349.6531, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(737.8197, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(469.6438, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(464.7685, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(717.9840, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(528.8417, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(456.9817, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(399.1115, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(211.9041, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(185.0807, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(496.8689, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(598.1974, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(414.1833, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(707.9371, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(514.4838, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(166.6325, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(212.6299, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(306.4687, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(379.6957, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(265.2627, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(727.8062, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(657.2528, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(899.7500, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(290.2178, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(350.8643, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(445.3036, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(443.1557, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(554.5875, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(195.3245, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(249.6755, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(217.1136, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(445.1569, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(608.2828, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(842.3798, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(500.6669, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(657.7776, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(370.6010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(572.2583, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(535.9206, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(561.5125, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(578.8565, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(800.6130, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(265.7145, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(268.4563, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(310.2042, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(577.2448, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(829.2374, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(403.3542, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(622.7991, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.1918158567774936, 'recall': 0.2798507462686567, 'f1': 0.2276176024279211, 'number': 268}, 'P': {'precision': 0.2360248447204969, 'recall': 0.33727810650887574, 'f1': 0.2777101096224117, 'number': 338}, 'overall_precision': 0.21624713958810068, 'overall_recall': 0.3118811881188119, 'overall_f1': 0.2554054054054054, 'overall_accuracy': 0.5564698735295742}
			------------EPOCH 10---------------
Loss:  tensor(847.7351, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(298.0965, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(215.9620, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(153.4650, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(283.7316, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(250.0167, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(518.6656, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(330.3498, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1073.7502, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(635.4623, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(317.1078, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(330.2762, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(641.9459, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(471.2845, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(413.6460, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(424.0754, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(234.7142, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(191.3826, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(610.8439, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(564.9819, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(398.4145, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(674.8702, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(676.2631, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(200.3662, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(226.1670, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(332.8901, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(403.6313, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(274.7106, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(408.1914, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(497.8306, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(792.0200, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(239.6674, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(266.6127, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(380.7114, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(390.4752, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(319.3287, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(175.4750, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(250.2280, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(194.4491, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(420.7785, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(330.7599, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(613.3441, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(408.8988, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(527.6997, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(319.8648, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(447.7119, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(531.6780, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(577.3702, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(470.5369, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(516.7167, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(216.2577, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(262.6129, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(210.5820, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(380.7131, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(486.6012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(379.6520, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(425.2731, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.17176470588235293, 'recall': 0.27238805970149255, 'f1': 0.21067821067821066, 'number': 268}, 'P': {'precision': 0.23809523809523808, 'recall': 0.23668639053254437, 'f1': 0.2373887240356083, 'number': 338}, 'overall_precision': 0.20105124835742444, 'overall_recall': 0.2524752475247525, 'overall_f1': 0.2238478419897586, 'overall_accuracy': 0.534544651239852}
			------------EPOCH 11---------------
Loss:  tensor(704.0471, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(275.4361, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(191.1741, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(111.2424, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(263.2057, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(203.4460, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(361.5196, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(286.9171, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(910.7575, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(622.3767, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(252.0047, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(291.3960, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(752.1558, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(528.3206, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(463.0073, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(427.9790, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(238.8653, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(200.6973, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(448.4894, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(706.3794, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(364.6070, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(678.0299, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(469.7902, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(128.2200, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(141.5727, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(250.6938, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(322.8517, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(163.8764, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(314.3170, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(289.0483, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(555.0343, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(166.8855, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(187.4359, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(233.9514, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(226.9284, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(180.9075, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(159.6240, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(194.6399, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(113.9840, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(315.4221, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(270.3135, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(407.7375, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(263.9945, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(466.0765, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(305.8662, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(510.9027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(486.3087, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(546.4328, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(480.3481, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(454.4113, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(314.7823, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(413.8568, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(448.8467, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(515.7192, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(667.8232, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(545.8118, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(590.1593, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.28502415458937197, 'recall': 0.22014925373134328, 'f1': 0.24842105263157893, 'number': 268}, 'P': {'precision': 0.20535714285714285, 'recall': 0.34023668639053256, 'f1': 0.2561247216035635, 'number': 338}, 'overall_precision': 0.22685788787483702, 'overall_recall': 0.2871287128712871, 'overall_f1': 0.2534595775673707, 'overall_accuracy': 0.6089909979565914}
			------------EPOCH 12---------------
Loss:  tensor(706.7178, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(235.4196, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(157.7999, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(94.4732, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(158.5099, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(91.4664, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(241.8857, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(152.3295, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(673.0297, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(378.8055, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(200.8419, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(221.8403, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(416.4394, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(308.2443, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(229.6500, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(212.9661, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(110.9975, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(117.5211, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(292.1214, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(386.1915, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(280.9662, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(546.7766, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(324.5657, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(80.5860, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(148.6425, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(236.2735, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(255.3620, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(212.0453, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(541.9429, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(448.5227, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(696.8254, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(247.8279, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(254.9826, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(297.5611, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(307.5922, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(353.0832, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(142.4523, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(120.1942, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(114.5496, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(338.3489, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(303.6047, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(410.8908, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(213.0265, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(424.7086, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(155.9632, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(153.7610, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(377.0394, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(262.7940, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(222.9327, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(403.3057, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(126.7920, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(107.7016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(128.4720, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(299.3922, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(426.6194, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(236.4684, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(301.5565, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.23497267759562843, 'recall': 0.16044776119402984, 'f1': 0.19068736141906872, 'number': 268}, 'P': {'precision': 0.15834767641996558, 'recall': 0.27218934911242604, 'f1': 0.20021762785636563, 'number': 338}, 'overall_precision': 0.17670157068062828, 'overall_recall': 0.22277227722772278, 'overall_f1': 0.19708029197080293, 'overall_accuracy': 0.6035234991992048}
			------------EPOCH 13---------------
Loss:  tensor(564.1942, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(154.4102, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(107.0606, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(65.4629, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(127.1253, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(86.9814, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(336.2449, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(178.9774, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(862.2545, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(485.2693, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(175.1472, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(246.1390, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(475.7399, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(337.7582, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(213.0459, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(264.2918, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(152.4996, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(116.1939, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(388.3873, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(366.5038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(242.2144, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(425.8736, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(259.6354, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(107.7265, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(119.9990, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(198.7451, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(145.6222, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(118.9979, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(245.1268, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(196.9009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(464.3279, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(117.4858, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(130.3384, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(149.8370, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(130.2119, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(94.9107, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(106.5229, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(82.8519, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.3992, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(196.7991, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(160.5458, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(209.7652, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(133.6614, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(233.8428, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(132.0525, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(131.5500, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(258.3440, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(194.1826, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(164.2947, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(328.4069, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(136.0985, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(97.6758, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(109.4169, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(270.4328, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(364.3312, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(168.8829, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(210.9179, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.17857142857142858, 'recall': 0.26119402985074625, 'f1': 0.2121212121212121, 'number': 268}, 'P': {'precision': 0.18778280542986425, 'recall': 0.2455621301775148, 'f1': 0.2128205128205128, 'number': 338}, 'overall_precision': 0.18345323741007194, 'overall_recall': 0.2524752475247525, 'overall_f1': 0.2125, 'overall_accuracy': 0.5675705528248743}
			------------EPOCH 14---------------
Loss:  tensor(485.8658, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(125.0533, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(78.6797, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.9794, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(84.9860, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(66.1710, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(149.3674, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(104.4782, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(412.7488, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(266.1232, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(119.9116, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(108.2074, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(293.3550, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(155.1641, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(132.7166, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(145.2710, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.9991, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(73.4137, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(202.8896, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(211.1724, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(173.9041, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(292.6752, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(119.2490, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.0000, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.4226, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(94.6981, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(77.4948, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(85.2984, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(195.0159, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(130.1942, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(346.3291, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(86.4553, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(89.2212, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(102.4670, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(81.8312, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.8036, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(87.5122, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.2663, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.4020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(139.6011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(117.5725, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(157.2701, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(90.5943, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(171.7242, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(76.6598, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(85.5974, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(207.8397, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(135.9014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(114.2025, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(197.2724, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(74.7844, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(59.4993, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(64.7987, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(174.2213, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(227.7591, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(103.3827, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(132.4128, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.23863636363636365, 'recall': 0.23507462686567165, 'f1': 0.2368421052631579, 'number': 268}, 'P': {'precision': 0.13559322033898305, 'recall': 0.21301775147928995, 'f1': 0.16570771001150747, 'number': 338}, 'overall_precision': 0.16981132075471697, 'overall_recall': 0.22277227722772278, 'overall_f1': 0.19271948608137046, 'overall_accuracy': 0.6008173634505992}
			------------EPOCH 15---------------
Loss:  tensor(451.6649, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.5169, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.4590, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.0175, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.7432, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.1198, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(113.9812, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.6714, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(308.2436, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(179.1747, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(78.7992, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(66.6858, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(227.7459, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(98.6889, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(70.3715, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(109.8993, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.7533, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.9305, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(148.4130, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(129.6836, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(128.6799, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(211.8784, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(107.4568, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.5008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.4086, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(70.8484, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.1698, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(62.0016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(166.9155, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(90.1094, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(275.0110, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.3775, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(69.5280, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(80.8131, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.7067, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.3885, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(72.1655, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.5626, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.7438, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(98.9849, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(96.2235, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(107.2854, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(65.9372, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(121.5840, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.2273, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.5156, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(192.6353, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(102.8401, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(84.3882, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(184.5750, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.0584, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.8534, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.8702, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(168.6030, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(240.6807, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(79.1908, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(106.4695, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.23024054982817868, 'recall': 0.25, 'f1': 0.23971377459749554, 'number': 268}, 'P': {'precision': 0.16827852998065765, 'recall': 0.257396449704142, 'f1': 0.20350877192982458, 'number': 338}, 'overall_precision': 0.1905940594059406, 'overall_recall': 0.25412541254125415, 'overall_f1': 0.21782178217821785, 'overall_accuracy': 0.590600320318109}
			------------EPOCH 16---------------
Loss:  tensor(419.2037, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.0534, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.5994, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.0304, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.5440, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.3452, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(79.5735, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.2103, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(246.2544, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(152.4275, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(59.4245, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.3812, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(206.4767, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(70.9448, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.3509, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(83.5051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.0509, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.7226, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(126.7184, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(94.6111, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(106.2941, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(177.1066, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(74.4332, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.4080, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.6292, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.1052, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.4043, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.0175, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(156.4153, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(66.8397, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(239.8716, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.6367, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.1904, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(61.4609, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.4402, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.2722, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(61.8376, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.1215, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.8043, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(68.6629, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.6118, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(80.7890, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.9756, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(85.1640, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.6947, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.3576, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(175.1813, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(83.4767, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(66.8337, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(158.2885, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.0504, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.6432, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.0403, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(125.8229, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(166.6930, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.2550, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(80.1465, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.2222222222222222, 'recall': 0.2537313432835821, 'f1': 0.23693379790940766, 'number': 268}, 'P': {'precision': 0.1488095238095238, 'recall': 0.22189349112426035, 'f1': 0.17814726840855108, 'number': 338}, 'overall_precision': 0.1765432098765432, 'overall_recall': 0.23597359735973597, 'overall_f1': 0.2019774011299435, 'overall_accuracy': 0.5921466836030265}
			------------EPOCH 17---------------
Loss:  tensor(300.9730, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.7666, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.2605, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.2339, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.4244, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.4870, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.4757, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.0516, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(209.7504, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(113.6564, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.0596, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.9234, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(178.9129, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(59.3796, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.9054, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.2332, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.9832, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.5327, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(118.4282, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(79.8614, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(88.5471, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(158.6180, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.7664, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.4203, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.1917, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.4088, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.4421, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.2179, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(125.7585, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.2092, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(186.3988, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.8420, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.6230, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.5061, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.3334, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.5004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.1388, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.1086, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.2397, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.6545, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.1611, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.6965, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.1354, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(64.4594, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.8413, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.2148, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(157.9459, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(62.9069, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.8385, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(119.7084, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.3070, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.0635, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.2895, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(106.5130, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(146.1765, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.1199, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(64.7330, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.2268041237113402, 'recall': 0.2462686567164179, 'f1': 0.23613595706618962, 'number': 268}, 'P': {'precision': 0.14065510597302505, 'recall': 0.21597633136094674, 'f1': 0.17036172695449242, 'number': 338}, 'overall_precision': 0.17160493827160495, 'overall_recall': 0.22937293729372937, 'overall_f1': 0.1963276836158192, 'overall_accuracy': 0.592367592643729}
			------------EPOCH 18---------------
Loss:  tensor(261.8996, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.5555, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.4836, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.7296, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.1689, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.8775, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.9398, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.0934, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(193.7934, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(84.2299, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.6839, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.5949, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(173.7070, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.8542, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.9134, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.6773, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.6260, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.5979, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(131.4165, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(84.6068, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(87.1090, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(157.9999, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.7965, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.0520, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.6372, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.1145, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.0633, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.7322, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(99.5863, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.9118, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(136.2437, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.8735, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.5785, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.7707, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.1601, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.2042, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.7394, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.0249, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.8988, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.8291, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.4795, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.4288, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.0822, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.2082, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.3803, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.5833, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(137.0676, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.7656, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.7065, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(117.1425, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.7428, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.2122, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.0223, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(90.5520, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(126.9592, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.5485, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.0569, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.22397476340694006, 'recall': 0.26492537313432835, 'f1': 0.24273504273504273, 'number': 268}, 'P': {'precision': 0.1656686626746507, 'recall': 0.2455621301775148, 'f1': 0.19785458879618595, 'number': 338}, 'overall_precision': 0.1882640586797066, 'overall_recall': 0.25412541254125415, 'overall_f1': 0.21629213483146068, 'overall_accuracy': 0.5893300933340697}
			------------EPOCH 19---------------
Loss:  tensor(249.2159, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.9884, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.1435, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.2049, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.5868, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.3324, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.9594, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.3751, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(170.0386, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(70.3546, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.2354, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.7539, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(164.2731, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.4918, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.8155, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.1604, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.7839, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.3226, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(116.4874, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(64.8179, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(67.3930, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(125.5694, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.9416, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.6143, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.1695, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.5014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.4539, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.3978, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(83.2272, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.7552, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(114.3831, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.5183, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.1524, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.6696, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.5465, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.2991, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.7854, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.2611, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.0219, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.7179, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.1419, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.5313, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.7603, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.6447, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.4731, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.4411, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(134.3181, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.4046, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.7719, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(124.7515, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.4932, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.8484, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.4536, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(87.4544, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(135.1482, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.5498, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.9386, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.22336769759450173, 'recall': 0.24253731343283583, 'f1': 0.23255813953488375, 'number': 268}, 'P': {'precision': 0.15428571428571428, 'recall': 0.23964497041420119, 'f1': 0.18771726535341832, 'number': 338}, 'overall_precision': 0.17892156862745098, 'overall_recall': 0.24092409240924093, 'overall_f1': 0.20534458509142053, 'overall_accuracy': 0.5917600927817971}
			------------EPOCH 20---------------
Loss:  tensor(217.7091, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.8611, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.5097, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.5513, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.6625, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.6208, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.7977, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.1054, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(157.5710, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.3923, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.5255, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.9496, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(119.8877, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.7908, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.6013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.0189, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.3104, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.4338, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(74.5745, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.6484, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.5008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(79.2194, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.2236, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.1065, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.7884, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.6030, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.8130, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.4787, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(121.7951, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.9335, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(97.7448, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.6288, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.7386, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.9149, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.3567, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.2995, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.3727, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.8106, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.5794, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.2044, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.6897, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.4286, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.8288, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.0805, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.3958, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.1290, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(153.4348, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.8021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.5131, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(113.8591, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.4282, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.6557, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.2796, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(103.9712, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(176.8711, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.4133, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.9348, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.22887323943661972, 'recall': 0.24253731343283583, 'f1': 0.23550724637681159, 'number': 268}, 'P': {'precision': 0.15471698113207547, 'recall': 0.24260355029585798, 'f1': 0.1889400921658986, 'number': 338}, 'overall_precision': 0.18058968058968058, 'overall_recall': 0.24257425742574257, 'overall_f1': 0.20704225352112673, 'overall_accuracy': 0.5884464571712598}
			------------EPOCH 21---------------
Loss:  tensor(194.7401, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.5277, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.4663, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.0520, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.2187, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.2652, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.8181, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.9431, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(139.1965, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.2765, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.5706, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.1897, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(128.1330, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.9409, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.1660, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.9478, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.8723, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.6909, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(86.6468, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.9098, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.3668, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(77.3436, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.9410, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.6396, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.6968, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.0738, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.9966, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.9180, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(109.8474, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.2749, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(96.9585, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.1131, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.1814, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.9658, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.2834, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.9741, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.7101, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.1643, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.6915, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.9614, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.6656, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.0492, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.2803, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.3014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.9107, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.7144, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(112.6322, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.2476, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.7947, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(93.5255, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.4196, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.1214, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.0308, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(81.1945, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(135.4704, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.6360, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.4219, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.22258064516129034, 'recall': 0.2574626865671642, 'f1': 0.23875432525951557, 'number': 268}, 'P': {'precision': 0.15682656826568267, 'recall': 0.2514792899408284, 'f1': 0.19318181818181818, 'number': 338}, 'overall_precision': 0.1807511737089202, 'overall_recall': 0.25412541254125415, 'overall_f1': 0.2112482853223594, 'overall_accuracy': 0.587231457447396}
			------------EPOCH 22---------------
Loss:  tensor(173.3013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.7164, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.2137, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.9834, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.2569, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.3488, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.0832, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.2774, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(127.4642, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.0514, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.1629, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.3302, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(119.9399, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.1288, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.1514, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.3135, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.3816, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.3392, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(72.2971, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.2074, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.2693, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(68.4869, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.7045, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.7100, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.9463, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.1529, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.6284, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.9868, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(59.9015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.5334, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(61.9933, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.0760, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.7273, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.2267, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.1508, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.2269, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.4055, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.8971, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.2225, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.3631, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.9555, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.2302, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.5402, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.2167, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.0294, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.3802, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(103.6506, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.8338, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.8472, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(81.9503, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.9401, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.9046, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.3100, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(64.5710, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(108.9994, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.6589, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.3581, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.21710526315789475, 'recall': 0.2462686567164179, 'f1': 0.23076923076923075, 'number': 268}, 'P': {'precision': 0.13307984790874525, 'recall': 0.20710059171597633, 'f1': 0.16203703703703706, 'number': 338}, 'overall_precision': 0.163855421686747, 'overall_recall': 0.22442244224422442, 'overall_f1': 0.18941504178272983, 'overall_accuracy': 0.5911525929198652}
			------------EPOCH 23---------------
Loss:  tensor(199.9602, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.0717, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.9464, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.7757, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.7695, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.0059, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.5727, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.4181, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(118.1016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.6134, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.9348, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.5472, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(99.4713, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.4201, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.6191, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.8282, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.3795, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.3737, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(61.7749, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.8639, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.9657, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.6941, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.6051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.8753, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.4455, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.7819, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.3252, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.1912, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(73.1447, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.3567, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(59.3331, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.9246, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.4114, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.7011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.0350, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.2798, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.8482, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.2141, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.7570, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.4543, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(59.6346, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.5292, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.2546, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.9686, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.4451, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.0950, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(103.6585, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.6763, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.3191, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(84.5473, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.5057, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.1228, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.0014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(64.5635, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(103.8367, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.4902, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.2251, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.21052631578947367, 'recall': 0.23880597014925373, 'f1': 0.22377622377622375, 'number': 268}, 'P': {'precision': 0.16044776119402984, 'recall': 0.25443786982248523, 'f1': 0.19679633867276888, 'number': 338}, 'overall_precision': 0.17857142857142858, 'overall_recall': 0.24752475247524752, 'overall_f1': 0.20746887966804978, 'overall_accuracy': 0.5841939581377368}
			------------EPOCH 24---------------
Loss:  tensor(290.9247, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.1006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.6389, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.6267, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.4184, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.9231, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.5441, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.1466, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(115.5434, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.8425, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.8999, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.0184, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(99.1083, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.8692, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.2959, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.3561, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.4849, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.8749, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(59.0610, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.1508, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.9353, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(66.2831, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.1728, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5278, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.6107, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.1864, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.0851, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.6328, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.9630, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.1619, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.0279, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.5168, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.9043, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.5696, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.3023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.3350, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.9888, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.6351, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.9754, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.3464, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.4389, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.4309, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.5151, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.7278, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.8712, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.7600, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(105.4547, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.1268, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.2023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(96.1053, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.0901, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.0314, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.1482, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(62.2628, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(106.7847, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.4999, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.8591, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.20552147239263804, 'recall': 0.25, 'f1': 0.2255892255892256, 'number': 268}, 'P': {'precision': 0.13181019332161686, 'recall': 0.22189349112426035, 'f1': 0.16538037486218302, 'number': 338}, 'overall_precision': 0.15865921787709497, 'overall_recall': 0.23432343234323433, 'overall_f1': 0.1892071952031979, 'overall_accuracy': 0.5867344121058155}
			------------EPOCH 25---------------
Loss:  tensor(253.3556, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.8040, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.6628, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.9113, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.6570, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.0745, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.0487, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.3016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(112.2626, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.9499, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.9795, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.8817, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(92.5534, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.5071, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.3042, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.7379, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.9460, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.7029, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.0645, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.9223, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.5564, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.3732, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.8565, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.1769, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.7437, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.7408, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.4473, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.9203, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.1221, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.9441, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.6816, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.8158, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.6259, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.1290, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.4033, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.4327, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.8335, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.2411, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.9725, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.3492, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.4849, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.8543, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.0115, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.2770, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(94.2495, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(185.9359, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(90.8440, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.1328, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.1536, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(94.5617, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.0292, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.7787, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.4551, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(117.3407, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(206.3824, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.6341, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.6829, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.20512820512820512, 'recall': 0.23880597014925373, 'f1': 0.2206896551724138, 'number': 268}, 'P': {'precision': 0.14772727272727273, 'recall': 0.23076923076923078, 'f1': 0.18013856812933027, 'number': 338}, 'overall_precision': 0.16904761904761906, 'overall_recall': 0.23432343234323433, 'overall_f1': 0.1964038727524205, 'overall_accuracy': 0.5882807753907329}
			------------EPOCH 26---------------
Loss:  tensor(187.4286, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.2263, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.6187, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.6029, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2244, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.4658, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.8997, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5699, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(100.9038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.5180, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.5212, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.7952, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(89.7049, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.8754, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.0438, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.1216, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.8095, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.2138, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.3068, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.1783, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.0981, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.1949, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.1242, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.0104, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.3698, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.5095, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.6405, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.4619, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.2784, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.9671, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.5436, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.6146, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.5053, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.7590, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.7478, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.3863, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.3323, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.2555, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.0953, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.5758, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.2936, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.4181, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.5315, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.9796, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.9981, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(69.6853, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(96.4886, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.7473, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.1374, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(78.6967, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.1120, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.7292, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.5410, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(66.0846, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(118.7387, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.8964, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.8207, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.2090032154340836, 'recall': 0.24253731343283583, 'f1': 0.22452504317789293, 'number': 268}, 'P': {'precision': 0.14234234234234233, 'recall': 0.23372781065088757, 'f1': 0.1769316909294513, 'number': 338}, 'overall_precision': 0.16628175519630484, 'overall_recall': 0.2376237623762376, 'overall_f1': 0.19565217391304346, 'overall_accuracy': 0.5906555475782846}
			------------EPOCH 27---------------
Loss:  tensor(173.9736, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.3722, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.0888, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.4039, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.0734, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.4843, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.8684, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.4565, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(116.2043, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.6473, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.9371, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.7458, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(79.0502, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.4479, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.0242, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.9199, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.3062, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.5425, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.6158, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.7575, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.7272, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.6643, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.9628, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.9943, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.0316, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.7961, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.8482, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.9164, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.0099, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.3655, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.9846, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.3999, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.4556, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.4517, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.0638, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5911, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.1045, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.5314, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.8111, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.4851, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.0372, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.8961, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.5552, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.8251, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.3260, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.4951, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(101.2760, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.0032, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.1440, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(69.1951, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2774, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.3846, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.1928, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.0984, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(82.2992, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.9386, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.6415, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.20743034055727555, 'recall': 0.25, 'f1': 0.22673434856175972, 'number': 268}, 'P': {'precision': 0.14181818181818182, 'recall': 0.23076923076923078, 'f1': 0.17567567567567569, 'number': 338}, 'overall_precision': 0.1660939289805269, 'overall_recall': 0.23927392739273928, 'overall_f1': 0.19607843137254902, 'overall_accuracy': 0.5917600927817971}
			------------EPOCH 28---------------
Loss:  tensor(151.7857, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.1781, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.1600, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.5654, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.0205, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.8380, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.4583, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.4902, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(157.9534, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.2750, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.5508, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.7163, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(75.9142, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.7051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.9192, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.6640, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.0954, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.3640, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.5463, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.9064, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.1141, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.1171, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.8362, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.8096, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.0103, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.7936, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.1773, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.3365, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.4546, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.1774, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.3245, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2543, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.8293, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.0177, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.1432, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.3534, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.7366, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.7491, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3081, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.9425, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.1963, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.3421, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.1379, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.6014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.0753, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.3011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(77.4662, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.4534, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.3501, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(66.0380, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.9177, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.1347, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.0256, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.5209, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(84.9441, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.2222, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.5660, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.1956521739130435, 'recall': 0.23507462686567165, 'f1': 0.21355932203389832, 'number': 268}, 'P': {'precision': 0.13394495412844037, 'recall': 0.21597633136094674, 'f1': 0.1653454133635334, 'number': 338}, 'overall_precision': 0.1568627450980392, 'overall_recall': 0.22442244224422442, 'overall_f1': 0.18465716225390358, 'overall_accuracy': 0.5839730490970343}
			------------EPOCH 29---------------
Loss:  tensor(142.8417, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.6827, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.3868, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.7507, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5339, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.0461, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.2911, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.0845, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(84.4031, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.9165, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.2866, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.0585, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(82.9543, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.4406, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.5633, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.7009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2491, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.9136, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.9454, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.5062, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.9670, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.1768, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.8420, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5104, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5732, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.6965, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3446, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.7982, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.3045, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.3681, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.4229, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.8340, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.4197, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.7844, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.4100, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.0275, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.0229, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.3018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2226, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.2417, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.9877, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.2194, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2042, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.6560, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.7078, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.8827, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(64.0104, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.1252, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.2581, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.3221, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5886, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6268, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6332, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.2063, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(77.8761, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.6408, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.2987, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.20512820512820512, 'recall': 0.23880597014925373, 'f1': 0.2206896551724138, 'number': 268}, 'P': {'precision': 0.12857142857142856, 'recall': 0.21301775147928995, 'f1': 0.1603563474387528, 'number': 338}, 'overall_precision': 0.1559633027522936, 'overall_recall': 0.22442244224422442, 'overall_f1': 0.18403247631935046, 'overall_accuracy': 0.5839730490970343}
			------------EPOCH 30---------------
Loss:  tensor(116.5921, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.1137, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.3179, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.7377, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4192, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4727, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.9796, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.8423, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(72.6729, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.3288, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.4250, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.3422, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(72.5104, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.6485, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.3895, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.6778, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6383, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.8276, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.1407, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.5018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.1869, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.2201, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.7317, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2606, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.3820, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.3790, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2476, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4837, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.9688, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.0192, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.7067, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5420, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.6887, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.2591, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.1148, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.7470, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.2887, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.0401, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0555, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.8411, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.7532, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.3179, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.0619, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.4959, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.6528, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.8535, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.7600, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.8049, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.6854, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.5456, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3733, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.8187, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5611, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.3728, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.3781, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.2285, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.3659, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.2, 'recall': 0.23880597014925373, 'f1': 0.217687074829932, 'number': 268}, 'P': {'precision': 0.1357142857142857, 'recall': 0.22485207100591717, 'f1': 0.16926503340757237, 'number': 338}, 'overall_precision': 0.1590909090909091, 'overall_recall': 0.23102310231023102, 'overall_f1': 0.18842530282637954, 'overall_accuracy': 0.5820400949908875}
	Train size: 50 Test size: 50


		-------------RUN 1-----------
			------------EPOCH 1---------------
Loss:  tensor(2751.3240, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2515.0449, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3641.6221, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2796.5103, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2394.7063, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1985.5640, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1847.5696, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1137.1790, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1339.6165, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1347.9790, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2275.1865, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3677.7832, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2351.8936, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3884.7083, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1495.6891, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1325.8932, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1247.3945, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1611.4016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2520.7661, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1482.5913, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1207.4323, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1616.0807, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2157.6018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1983.4006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2144.9521, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2421.7163, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2206.2917, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1259.8579, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2570.0210, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1659.0215, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1454.6617, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2948.1636, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2132.4146, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3173.4839, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2505.6992, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3852.5840, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1849.7953, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1371.8547, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2362.8906, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.0036036036036036037, 'recall': 0.0029282576866764276, 'f1': 0.003231017770597739, 'number': 683}, 'P': {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'number': 839}, 'overall_precision': 0.0012106537530266344, 'overall_recall': 0.001314060446780552, 'overall_f1': 0.001260239445494644, 'overall_accuracy': 0.3703721722208708}
			------------EPOCH 2---------------
Loss:  tensor(2109.3420, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1953.8712, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2858.2378, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2092.8611, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2011.9570, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1767.8350, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1562.3026, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(897.9695, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1181.5093, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1147.6107, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2003.6740, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3269.2578, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2022.7073, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3569.4194, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1308.5994, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1203.1974, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1112.8162, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1410.7310, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2218.0542, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1291.5178, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1035.5369, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1394.0795, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1965.4553, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1792.1233, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2006.4274, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2228.7266, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2012.6742, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1144.2346, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2354.6230, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1549.3904, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1296.3684, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2712.5139, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2086.0383, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3060.3560, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2332.6777, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3657.7131, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1832.9287, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1311.1094, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2398.5073, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.052873563218390804, 'recall': 0.06734992679355783, 'f1': 0.0592401802962009, 'number': 683}, 'P': {'precision': 0.006731488406881077, 'recall': 0.010727056019070322, 'f1': 0.008272058823529412, 'number': 839}, 'overall_precision': 0.02492070684186679, 'overall_recall': 0.03613666228646518, 'overall_f1': 0.029498525073746312, 'overall_accuracy': 0.3962539528095354}
			------------EPOCH 3---------------
Loss:  tensor(1794.9788, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1968.1726, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2522.7632, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2025.0911, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1915.3120, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1467.8843, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1358.7723, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(807.3271, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1060.8333, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1095.4869, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1846.5348, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2608.0400, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1690.1792, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3160.7969, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1168.4767, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1114.2280, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(993.4445, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1277.5948, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2060.5845, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1169.1187, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(910.3506, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1246.8975, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1718.2910, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1677.0889, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1888.3840, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2017.3562, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1758.1658, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(994.7224, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2049.1262, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1399.8959, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1156.9390, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2445.6646, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1624.3314, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2217.6479, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2120.5757, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2951.7710, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1347.4391, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1099.3683, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1738.8193, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.044906900328587074, 'recall': 0.060029282576866766, 'f1': 0.05137844611528822, 'number': 683}, 'P': {'precision': 0.01327433628318584, 'recall': 0.028605482717520857, 'f1': 0.018133736305251228, 'number': 839}, 'overall_precision': 0.02388827636898199, 'overall_recall': 0.042706964520367936, 'overall_f1': 0.030638699033702567, 'overall_accuracy': 0.5011675991243006}
			------------EPOCH 4---------------
Loss:  tensor(1507.5771, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1688.9513, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2136.7866, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1604.3073, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1638.9979, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1366.4893, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1114.4122, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(675.9901, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(917.0343, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(940.4683, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1650.3054, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2148.2175, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1357.4536, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2717.2490, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1000.4754, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(981.6436, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(814.2975, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1129.4851, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1874.4899, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1009.2671, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(695.0171, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1093.8738, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1478.3926, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1423.9419, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1593.4895, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1743.7949, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1430.2166, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(755.2808, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1622.0413, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1312.8512, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(980.6468, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2124.2129, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1435.9858, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1922.7454, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1878.0525, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2462.7344, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1019.7917, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(888.9026, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1249.7576, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.08596837944664032, 'recall': 0.1273792093704246, 'f1': 0.10265486725663717, 'number': 683}, 'P': {'precision': 0.02095238095238095, 'recall': 0.03933253873659118, 'f1': 0.027340513670256836, 'number': 839}, 'overall_precision': 0.04638577502899111, 'overall_recall': 0.07884362680683311, 'overall_f1': 0.058408371866634216, 'overall_accuracy': 0.5383118462661153}
			------------EPOCH 5---------------
Loss:  tensor(1158.8568, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1414.0431, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1825.1686, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1375.7544, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1459.7080, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1149.6619, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(857.7766, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(539.1620, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(698.4337, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(680.5564, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1506.0457, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1703.9624, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1008.3898, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2243.6538, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(921.0489, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(844.9979, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(631.1947, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1006.8254, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1767.1260, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(827.4525, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(516.2354, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(898.7119, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1237.0173, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1189.7041, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1286.1880, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1464.5278, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1185.0895, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(618.1981, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1317.9294, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1152.8989, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(834.0223, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1802.9133, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1103.5103, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1458.4546, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1630.4377, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1889.1765, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(687.1439, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(719.4641, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(788.0531, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.13302217036172695, 'recall': 0.16691068814055637, 'f1': 0.14805194805194805, 'number': 683}, 'P': {'precision': 0.06437768240343347, 'recall': 0.12514898688915377, 'f1': 0.08502024291497975, 'number': 839}, 'overall_precision': 0.0880225080385852, 'overall_recall': 0.14388961892247043, 'overall_f1': 0.10922693266832917, 'overall_accuracy': 0.5720992459255656}
			------------EPOCH 6---------------
Loss:  tensor(841.5412, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1063.7991, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1498.6909, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1241.6344, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1348.5304, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(833.8685, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(527.6453, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(366.7147, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(400.8064, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(491.0323, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1274.8179, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1231.2478, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(810.8611, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1771.0381, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(750.5474, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(643.8636, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(489.0682, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(859.6516, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1437.3542, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(636.0934, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(389.7255, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(706.0693, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(967.2230, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(934.5757, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(962.8080, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1190.3765, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(950.3164, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(500.9196, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(972.2628, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(891.1935, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(488.4875, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1297.6055, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(842.8577, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(936.8929, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1171.1121, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1304.1438, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(457.3500, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(497.6364, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(534.4089, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.18127147766323023, 'recall': 0.3089311859443631, 'f1': 0.2284786139685977, 'number': 683}, 'P': {'precision': 0.06499535747446611, 'recall': 0.08343265792610251, 'f1': 0.07306889352818373, 'number': 839}, 'overall_precision': 0.1253904506916555, 'overall_recall': 0.18462549277266754, 'overall_f1': 0.14934892373106565, 'overall_accuracy': 0.5569934322549258}
			------------EPOCH 7---------------
Loss:  tensor(660.7216, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(827.2267, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1336.4294, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1243.3005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1038.3977, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(529.9298, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(385.7569, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(260.9208, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(281.3649, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(431.5728, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1742.7689, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1560.5366, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(719.0370, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1398.7854, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(731.5173, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(532.1700, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(370.7292, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(689.1863, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1252.9789, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(517.0742, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(422.7380, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(677.4621, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(863.0657, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1193.8127, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1241.5479, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1353.4227, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1106.4702, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(511.4304, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(776.9811, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(732.6603, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(502.8926, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1161.9481, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(652.5035, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(661.4877, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(836.7753, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(960.4365, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(405.0593, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(448.5030, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(473.5914, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.19247467438494936, 'recall': 0.19472913616398244, 'f1': 0.1935953420669578, 'number': 683}, 'P': {'precision': 0.09533267130089375, 'recall': 0.22884386174016685, 'f1': 0.13459516298633017, 'number': 839}, 'overall_precision': 0.12014787430683918, 'overall_recall': 0.21353482260183967, 'overall_f1': 0.15377336172226164, 'overall_accuracy': 0.5683775237168572}
			------------EPOCH 8---------------
Loss:  tensor(1012.0093, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(829.4330, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1567.2601, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1238.6863, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1130.3199, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(491.7604, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(402.2531, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(268.1635, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(295.1122, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(448.4860, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1164.4722, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1194.8306, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(686.6908, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1110.3673, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(551.0126, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(604.6523, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(403.7393, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(598.7540, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1210.6112, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(525.7595, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(257.5103, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(547.8796, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(821.0555, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(562.7848, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(617.1058, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(890.1505, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(668.7128, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(325.3637, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(610.6219, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(558.9468, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(328.4703, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(851.9708, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(629.9084, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(510.5770, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(781.5397, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(969.8286, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(310.5519, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(354.9594, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(452.0998, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.1423487544483986, 'recall': 0.2342606149341142, 'f1': 0.17708909795240735, 'number': 683}, 'P': {'precision': 0.04797353184449959, 'recall': 0.06912991656734208, 'f1': 0.05664062499999999, 'number': 839}, 'overall_precision': 0.09344192027432491, 'overall_recall': 0.14323258869908015, 'overall_f1': 0.11309987029831388, 'overall_accuracy': 0.5762344928241304}
			------------EPOCH 9---------------
Loss:  tensor(501.5048, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(732.5209, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1096.8071, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1064.0701, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(860.0288, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(446.7115, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(267.5612, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(208.1469, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(177.4437, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(253.9911, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(850.3336, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(890.8374, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(411.3900, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(701.6181, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(408.8746, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(359.2205, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(254.7686, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(523.4939, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1238.5393, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(401.8838, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(250.3643, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(579.0897, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(874.7861, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(507.7849, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(637.4484, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(967.0176, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(672.5145, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(294.3016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(453.1920, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(494.1421, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(270.1119, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(711.9849, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(544.5648, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(442.1655, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(507.6482, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(618.8252, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(224.0225, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(254.4043, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(269.4517, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.16631280962491155, 'recall': 0.3440702781844802, 'f1': 0.22423664122137404, 'number': 683}, 'P': {'precision': 0.03860294117647059, 'recall': 0.050059594755661505, 'f1': 0.04359107420861443, 'number': 839}, 'overall_precision': 0.11075569772091164, 'overall_recall': 0.18199737187910645, 'overall_f1': 0.13770817797663437, 'overall_accuracy': 0.5433714424714181}
			------------EPOCH 10---------------
Loss:  tensor(359.2648, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(523.1084, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(721.2262, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(685.4404, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(734.2612, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(463.0157, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(292.7610, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(240.8428, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(214.4560, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(381.6856, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(704.0317, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(748.0353, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(465.4228, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(861.0034, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(476.1410, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(411.6714, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(179.6060, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(346.3387, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(908.6873, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(271.0454, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(180.6582, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(357.2403, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(444.6081, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(368.9500, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(437.8515, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(674.1743, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(421.9996, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(234.1736, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(551.3522, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(458.6158, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(234.6046, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(639.6603, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(409.8174, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(544.6335, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(668.3722, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(551.1657, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(225.9351, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(169.8723, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(205.4052, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.19347496206373294, 'recall': 0.3733528550512445, 'f1': 0.25487256371814093, 'number': 683}, 'P': {'precision': 0.11261261261261261, 'recall': 0.11918951132300358, 'f1': 0.11580775911986103, 'number': 839}, 'overall_precision': 0.16092475067996373, 'overall_recall': 0.23324572930354795, 'overall_f1': 0.19045064377682402, 'overall_accuracy': 0.5455606908294819}
			------------EPOCH 11---------------
Loss:  tensor(346.4686, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(347.9329, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(631.4187, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(707.2068, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(716.2966, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(305.3902, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(244.0947, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(149.5060, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(163.2167, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(265.5596, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(866.3911, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(919.3730, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(495.8169, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1036.2896, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(383.0354, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(449.1756, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(267.8302, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(421.3073, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(903.7211, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(333.9378, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(144.8382, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(294.2063, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(354.0186, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(333.5309, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(334.4549, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(585.0183, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(352.2021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(192.7293, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(346.4930, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(328.5801, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(144.7830, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(517.3336, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(417.4136, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(320.4988, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(415.4212, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(732.4087, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(263.3052, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(337.4621, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(313.7722, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.1867881548974943, 'recall': 0.12005856515373353, 'f1': 0.14616755793226383, 'number': 683}, 'P': {'precision': 0.09722897423432182, 'recall': 0.23837902264600716, 'f1': 0.13812154696132597, 'number': 839}, 'overall_precision': 0.11298076923076923, 'overall_recall': 0.18528252299605782, 'overall_f1': 0.14036834245893479, 'overall_accuracy': 0.598808075893943}
			------------EPOCH 12---------------
Loss:  tensor(568.9211, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(602.3309, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(981.8510, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1092.6721, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(811.6434, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(290.4281, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(202.0932, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(157.8569, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(134.6285, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(178.5227, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(775.3008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(813.9125, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(314.8878, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(517.6772, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(315.2570, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(241.4255, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(114.5845, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(255.9967, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(670.8560, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(179.0726, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(131.2663, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(308.0345, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(411.2767, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(300.5616, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(485.1705, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(816.3054, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(496.5264, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(208.1413, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(450.9561, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(532.2705, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(264.2440, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(756.9802, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(428.0806, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(320.5891, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(456.8127, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(690.5692, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(151.2757, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(121.7126, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(146.2106, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.2, 'recall': 0.3411420204978038, 'f1': 0.25216450216450215, 'number': 683}, 'P': {'precision': 0.08313155770782889, 'recall': 0.12276519666269368, 'f1': 0.09913378248315688, 'number': 839}, 'overall_precision': 0.13976705490848584, 'overall_recall': 0.22076215505913271, 'overall_f1': 0.17116658176260824, 'overall_accuracy': 0.5994162004378497}
			------------EPOCH 13---------------
Loss:  tensor(278.2982, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(228.4332, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(404.1287, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(459.8863, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(370.4001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(197.9187, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(108.8763, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(105.2776, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(74.7496, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(117.8950, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(480.9892, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(481.0083, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(256.1110, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(459.9333, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(341.6967, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(236.3605, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(99.5644, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(192.6253, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(713.7065, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(195.4740, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(110.0809, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(249.7147, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(298.4466, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(223.6402, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(284.0189, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(471.1496, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(257.0430, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(137.4654, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(235.0764, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(239.2659, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(85.3242, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(333.8340, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(186.6122, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(181.6877, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(151.7373, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(356.7301, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(133.9455, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(123.1096, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(132.3651, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.18417945690672963, 'recall': 0.22840409956076135, 'f1': 0.20392156862745098, 'number': 683}, 'P': {'precision': 0.08260105448154657, 'recall': 0.16805721096543505, 'f1': 0.11076197957580519, 'number': 839}, 'overall_precision': 0.11628817541111981, 'overall_recall': 0.19513797634691196, 'overall_f1': 0.14573110893032384, 'overall_accuracy': 0.5986864509851617}
			------------EPOCH 14---------------
Loss:  tensor(183.7473, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(199.1938, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(294.5820, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(393.7307, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(289.0047, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(162.2808, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(76.2860, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(85.2312, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.6778, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(77.9887, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(383.2690, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(421.5966, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(201.6257, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(363.1725, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(213.8335, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(181.2226, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(64.9080, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(155.1045, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(496.0555, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(112.9880, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.1751, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(125.4404, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(168.5565, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(138.3094, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(159.5169, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(415.6671, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(181.2783, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(84.3304, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(158.1521, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(209.8604, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(65.7770, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(257.7837, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(136.3447, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(160.1619, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(108.0083, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(314.9293, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(106.5750, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(65.5302, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(85.4325, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.20570866141732283, 'recall': 0.30600292825768666, 'f1': 0.24602707474985286, 'number': 683}, 'P': {'precision': 0.09876543209876543, 'recall': 0.16209773539928488, 'f1': 0.12274368231046931, 'number': 839}, 'overall_precision': 0.14417049728374426, 'overall_recall': 0.2266754270696452, 'overall_f1': 0.17624521072796934, 'overall_accuracy': 0.6048163463877402}
			------------EPOCH 15---------------
Loss:  tensor(152.6929, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(146.0157, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(225.8607, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(319.7606, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(229.5770, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(110.7533, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.4485, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(67.5680, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.4662, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.1540, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(337.3796, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(360.1947, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(142.7547, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(283.5980, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(159.4756, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(143.1754, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.6731, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(102.6927, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(413.3969, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(82.0171, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.0153, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(99.2178, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(126.9987, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(108.0617, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(118.0332, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(349.8965, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(127.7521, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(65.2463, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(137.0015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(152.5668, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.6702, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(221.4598, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(98.8703, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(101.7268, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(74.6486, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(258.3788, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(83.2173, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.8248, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(73.0888, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.20327102803738317, 'recall': 0.2547584187408492, 'f1': 0.22612085769980508, 'number': 683}, 'P': {'precision': 0.13206262763784887, 'recall': 0.23122765196662692, 'f1': 0.1681109185441941, 'number': 839}, 'overall_precision': 0.15827956989247313, 'overall_recall': 0.24178712220762155, 'overall_f1': 0.19131791005978685, 'overall_accuracy': 0.6142544393091706}
			------------EPOCH 16---------------
Loss:  tensor(117.9669, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(126.0152, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(183.1402, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(268.9339, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(186.0341, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(85.7388, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.4528, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.4913, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.3422, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.3467, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(291.4692, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(324.3410, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(113.1085, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(212.2383, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(118.7951, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(110.2022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.0031, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(80.0499, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(336.4396, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(64.0679, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.6350, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(78.4169, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(91.9415, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(79.3941, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(84.5489, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(304.6959, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(100.7579, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.1891, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(99.6357, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(134.2950, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.7931, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(176.2988, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(71.6050, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(74.7388, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.8717, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(235.0726, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(62.3036, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.9779, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.4734, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.2046659597030753, 'recall': 0.28257686676427524, 'f1': 0.23739237392373924, 'number': 683}, 'P': {'precision': 0.11843971631205674, 'recall': 0.19904648390941598, 'f1': 0.14851044908848376, 'number': 839}, 'overall_precision': 0.1529961750956226, 'overall_recall': 0.23653088042049936, 'overall_f1': 0.18580645161290318, 'overall_accuracy': 0.6113597664801751}
			------------EPOCH 17---------------
Loss:  tensor(98.6824, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(96.7955, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(155.3425, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(225.1346, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(162.3450, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.1922, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.5758, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.8220, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.1530, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.0750, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(260.3492, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(296.8464, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(84.0364, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(161.8786, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(95.0353, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(84.8191, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.0235, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.7281, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(279.4990, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.8583, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.4488, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(64.7019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(69.4966, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.4762, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(66.7018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(268.3793, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(81.4472, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.1448, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(80.2444, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(118.5148, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.2474, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(141.2430, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.9878, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.9276, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.2962, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(216.7272, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.7104, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.9045, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.9994, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.20387243735763097, 'recall': 0.26207906295754024, 'f1': 0.2293401665598975, 'number': 683}, 'P': {'precision': 0.14285714285714285, 'recall': 0.2431466030989273, 'f1': 0.1799735333039259, 'number': 839}, 'overall_precision': 0.16608846487424112, 'overall_recall': 0.2516425755584757, 'overall_f1': 0.2001044932079415, 'overall_accuracy': 0.6121624908781318}
			------------EPOCH 18---------------
Loss:  tensor(81.8191, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(79.3382, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(131.0251, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(191.6842, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(131.5063, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.9456, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.6006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.6951, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.9636, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.9135, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(230.1917, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(271.0022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.5902, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(121.0146, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(74.6811, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(66.4289, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.3856, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.3905, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(228.1009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.2794, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.0352, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.5572, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.4739, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.5096, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.6357, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(236.3447, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(66.4276, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.5545, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.9254, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(109.6316, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.7984, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(116.7905, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.0534, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.0745, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.8575, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(201.5079, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.0059, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.6455, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.9449, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.20643729189789123, 'recall': 0.27232796486090777, 'f1': 0.23484848484848486, 'number': 683}, 'P': {'precision': 0.1457000710732054, 'recall': 0.24433849821215733, 'f1': 0.182546749777382, 'number': 839}, 'overall_precision': 0.16941074523396882, 'overall_recall': 0.2568988173455979, 'overall_f1': 0.204177545691906, 'overall_accuracy': 0.6108975918268061}
			------------EPOCH 19---------------
Loss:  tensor(68.9007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(64.0192, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(112.3852, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(162.3628, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(112.0952, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.7210, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.4422, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.9903, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.2403, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.4941, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(204.8352, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(246.8920, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.5733, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(91.2191, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.3584, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.0079, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.0008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.7069, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(188.8593, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.2371, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.4981, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.5329, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.4937, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.5655, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.2844, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(210.2377, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.3651, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.2526, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.3704, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(108.6215, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.3557, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(105.5460, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.3957, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.2380, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.5023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(189.5603, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.7446, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.8007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.6702, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.1984732824427481, 'recall': 0.2664714494875549, 'f1': 0.22749999999999998, 'number': 683}, 'P': {'precision': 0.14583333333333334, 'recall': 0.24195470798569726, 'f1': 0.18198117436127298, 'number': 839}, 'overall_precision': 0.1667388479861412, 'overall_recall': 0.25295663600525625, 'overall_f1': 0.20099190811798487, 'overall_accuracy': 0.6095597178302117}
			------------EPOCH 20---------------
Loss:  tensor(54.0472, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.5705, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(93.2311, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(137.2374, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(92.5671, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.9221, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.6259, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.6726, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.0083, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.6046, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(186.5813, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(231.3209, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.8576, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(71.8222, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.7614, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.6532, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.6808, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.1142, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(162.1195, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.6757, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.0410, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.7048, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.9632, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.5134, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.4385, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(189.1657, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.0481, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.3006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.5237, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(119.7091, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.9822, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(75.7532, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.8913, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.9879, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.3673, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(178.3020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.2295, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.9096, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.3982, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.21682464454976302, 'recall': 0.2679355783308931, 'f1': 0.23968565815324164, 'number': 683}, 'P': {'precision': 0.14596491228070174, 'recall': 0.24791418355184744, 'f1': 0.18374558303886926, 'number': 839}, 'overall_precision': 0.17232260907888938, 'overall_recall': 0.2568988173455979, 'overall_f1': 0.2062780269058296, 'overall_accuracy': 0.6072731695451229}
			------------EPOCH 21---------------
Loss:  tensor(43.5289, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.4932, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(83.8461, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(128.3919, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(77.3287, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.8524, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.5562, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.9598, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.7520, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.9058, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(185.2706, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(239.5850, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.1323, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(61.6251, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.2977, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.7591, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.9271, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.1777, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(139.4640, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.3705, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.0156, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.3751, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.3774, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.7381, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.8641, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(162.2667, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.6289, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.2211, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.6378, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(77.1551, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.5195, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.6856, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.4039, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.2248, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.9486, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(169.5169, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.3594, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.1826, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.0842, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.21255506607929514, 'recall': 0.28257686676427524, 'f1': 0.24261470773098678, 'number': 683}, 'P': {'precision': 0.14387464387464388, 'recall': 0.24076281287246723, 'f1': 0.18011591618368256, 'number': 839}, 'overall_precision': 0.1708477508650519, 'overall_recall': 0.259526938239159, 'overall_f1': 0.20605112154407926, 'overall_accuracy': 0.6073218195086354}
			------------EPOCH 22---------------
Loss:  tensor(45.0923, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.4775, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(108.1197, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(172.1122, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(75.4256, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.1353, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.5817, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.1463, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.7860, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.8542, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(191.6263, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(256.4626, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.3516, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.1136, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.1705, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.0773, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.6194, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.6911, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(124.4495, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.6275, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.5362, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.8162, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.1770, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.2071, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.4656, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(150.9987, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.6375, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.5587, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.5440, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(82.6669, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.0242, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.4131, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.7460, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.7932, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.4469, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(164.4465, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.5754, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.3500, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.1337, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.2119760479041916, 'recall': 0.2591508052708638, 'f1': 0.23320158102766797, 'number': 683}, 'P': {'precision': 0.1483050847457627, 'recall': 0.25029797377830754, 'f1': 0.18625277161862527, 'number': 839}, 'overall_precision': 0.17192358951577077, 'overall_recall': 0.2542706964520368, 'overall_f1': 0.20514179697853166, 'overall_accuracy': 0.6021892483580638}
			------------EPOCH 23---------------
Loss:  tensor(122.9068, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.9598, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(170.0099, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(219.5007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(93.7650, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.2209, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.8267, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.0941, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.8052, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.6006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(206.7339, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(312.7033, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(117.3289, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(81.7707, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.3270, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.1359, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.2469, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.6162, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(154.2650, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.9837, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.4617, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.5752, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.9075, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.7240, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.1667, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(140.6973, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.7172, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.8081, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.1068, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(79.9355, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.1787, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(59.1981, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.1035, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.0792, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.7890, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(169.8347, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.9118, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.8412, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.5401, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.20106951871657755, 'recall': 0.2752562225475842, 'f1': 0.23238566131025962, 'number': 683}, 'P': {'precision': 0.12627551020408162, 'recall': 0.2359952324195471, 'f1': 0.16452014956377234, 'number': 839}, 'overall_precision': 0.15421494206951658, 'overall_recall': 0.2536136662286465, 'overall_f1': 0.19180124223602485, 'overall_accuracy': 0.607005594745804}
			------------EPOCH 24---------------
Loss:  tensor(83.2819, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.2522, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(138.2454, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(197.5627, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(92.3246, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.1815, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.3122, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.3449, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.4455, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.9726, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(171.7187, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(241.5491, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(168.0105, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(108.1790, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.2412, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.2956, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.3918, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.3501, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(177.1609, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.3163, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.9953, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.8512, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.1372, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.1141, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.8689, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(191.6355, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.5038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.6605, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.3138, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(71.5457, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.6777, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.1879, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.2584, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.9570, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.1293, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(163.1617, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.8975, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.4104, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.9018, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.20303383897316218, 'recall': 0.2547584187408492, 'f1': 0.22597402597402597, 'number': 683}, 'P': {'precision': 0.1338818249813014, 'recall': 0.2133492252681764, 'f1': 0.1645220588235294, 'number': 839}, 'overall_precision': 0.1608933454876937, 'overall_recall': 0.23193166885676741, 'overall_f1': 0.18998923573735196, 'overall_accuracy': 0.6104110921916809}
			------------EPOCH 25---------------
Loss:  tensor(63.8240, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(64.9599, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(123.7374, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(165.9719, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(98.3438, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.9578, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.5210, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.2067, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.4796, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.9426, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(152.0130, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(211.8714, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(78.0037, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.3677, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.0245, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.5034, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.6066, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.4748, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(123.3659, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.8697, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.9910, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.6401, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.8457, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.0849, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.0347, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(149.8527, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.7135, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.7306, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.5441, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(78.9512, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.2883, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.3887, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.8316, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.6987, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.5355, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(160.2968, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.5166, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.8643, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.8521, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.22125813449023862, 'recall': 0.2986822840409956, 'f1': 0.25420560747663556, 'number': 683}, 'P': {'precision': 0.1619718309859155, 'recall': 0.27413587604290823, 'f1': 0.2036299247454626, 'number': 839}, 'overall_precision': 0.18531169940222034, 'overall_recall': 0.2851511169513798, 'overall_f1': 0.22463768115942034, 'overall_accuracy': 0.605351495986378}
			------------EPOCH 26---------------
Loss:  tensor(107.2875, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.5645, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(125.3016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(138.0823, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(65.3517, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.0138, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.6537, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.7320, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.7418, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.7495, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(126.5791, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(182.5632, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.9220, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.3608, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.3945, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.5695, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.0508, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.4551, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(119.1452, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.0862, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.2810, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.2154, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.9211, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.6623, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.9833, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(145.7358, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.6637, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.7644, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.4638, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(70.7138, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.9101, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.8446, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.4539, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.2040, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.0942, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(155.3406, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.5160, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.2023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.2722, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.24053452115812918, 'recall': 0.3162518301610542, 'f1': 0.2732447817836813, 'number': 683}, 'P': {'precision': 0.161698956780924, 'recall': 0.2586412395709178, 'f1': 0.19899128839981658, 'number': 839}, 'overall_precision': 0.19330357142857144, 'overall_recall': 0.2844940867279895, 'overall_f1': 0.2301967038809144, 'overall_accuracy': 0.60462174653369}
			------------EPOCH 27---------------
Loss:  tensor(79.8398, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.9611, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(122.0073, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(157.3282, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(64.3071, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.1722, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.2271, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.8427, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.9936, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.7889, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(119.4385, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(173.5738, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.6461, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.2241, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.7928, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.5747, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.4308, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.3570, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(92.3440, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.6139, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.6208, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.7110, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.7236, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.7910, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.9695, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(125.0903, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.5188, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.8055, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.7402, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(70.7255, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.6009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.7244, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.5149, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.1093, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.9483, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(161.0014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.5362, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.1863, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.4905, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.193446088794926, 'recall': 0.2679355783308931, 'f1': 0.22467771639042358, 'number': 683}, 'P': {'precision': 0.1401673640167364, 'recall': 0.2395709177592372, 'f1': 0.17685877694676638, 'number': 839}, 'overall_precision': 0.16134453781512606, 'overall_recall': 0.25229960578186594, 'overall_f1': 0.19682214249103025, 'overall_accuracy': 0.6111165166626125}
			------------EPOCH 28---------------
Loss:  tensor(72.1488, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.3265, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(89.1445, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(106.9025, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.9793, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.3099, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.5578, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.8832, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.1285, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.3755, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(105.1338, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(157.2137, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.5807, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.4394, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.0690, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.8894, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.1476, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.1081, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(87.3190, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.1677, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.4796, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.2986, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.2767, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.6079, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.8316, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(153.2168, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.7375, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.0736, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.5106, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(78.4435, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.5882, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.5315, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.0796, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.1463, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.1655, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(137.9692, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.2537, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.0889, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.1163, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.20855614973262032, 'recall': 0.28550512445095166, 'f1': 0.24103831891223731, 'number': 683}, 'P': {'precision': 0.15544041450777202, 'recall': 0.25029797377830754, 'f1': 0.19178082191780824, 'number': 839}, 'overall_precision': 0.17716535433070865, 'overall_recall': 0.26609724047306177, 'overall_f1': 0.21271008403361344, 'overall_accuracy': 0.6094380929214304}
			------------EPOCH 29---------------
Loss:  tensor(42.2957, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.0718, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(61.3969, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(80.4457, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.8097, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.7993, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.5558, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.3930, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.5898, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.1023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(95.3945, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(148.9113, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.8485, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.2919, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.3149, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.2197, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.2132, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.1806, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(78.7518, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.7112, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.3948, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.2988, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.2654, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.1868, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.6694, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(125.1648, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.4627, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.4262, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.9169, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(84.2586, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5203, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.2256, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.0977, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.8982, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.4211, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(129.3032, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.5741, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.0589, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.0580, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.21390374331550802, 'recall': 0.29282576866764276, 'f1': 0.24721878862793573, 'number': 683}, 'P': {'precision': 0.15838068181818182, 'recall': 0.265792610250298, 'f1': 0.19848687138406765, 'number': 839}, 'overall_precision': 0.18053777208706787, 'overall_recall': 0.2779237844940867, 'overall_f1': 0.21888745148771022, 'overall_accuracy': 0.6099002675747993}
			------------EPOCH 30---------------
Loss:  tensor(28.4349, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.1181, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.6384, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(59.3323, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.5756, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.7659, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.9100, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.3503, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.9896, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.7258, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(87.7544, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(137.1621, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.7682, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.3349, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.6547, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.7032, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.1952, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.2837, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(70.8743, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.1726, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.8242, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.8739, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.1319, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.8625, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.2518, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(84.6616, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.3675, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.3376, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.2917, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(62.7046, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.0780, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.7852, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.4336, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.6840, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.0704, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(115.9547, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.1030, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.5747, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.5173, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.20995670995670995, 'recall': 0.2840409956076135, 'f1': 0.24144368388301182, 'number': 683}, 'P': {'precision': 0.164079822616408, 'recall': 0.26460071513706795, 'f1': 0.20255474452554745, 'number': 839}, 'overall_precision': 0.18269653052261747, 'overall_recall': 0.2733245729303548, 'overall_f1': 0.21900500131613582, 'overall_accuracy': 0.6093164680126489}


		-------------RUN 2-----------
			------------EPOCH 1---------------
Loss:  tensor(3304.7437, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1908.4930, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2932.9561, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2153.4924, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1278.7043, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1345.2593, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3277.7676, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2514.2900, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2018.1719, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(612.6207, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1228.5337, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2270.2297, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1311.3845, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2674.7363, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2963.4648, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2335.2588, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1938.6498, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1509.4038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2636.8730, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(932.1072, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2023.1290, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1620.3420, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1750.1396, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2203.9282, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2507.4316, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2248.6772, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2170.5723, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1938.8247, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1439.9557, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1875.5045, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1549.0071, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1589.3259, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2065.2803, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1320.8669, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1232.4242, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.012531328320802004, 'recall': 0.006273525721455458, 'f1': 0.008361204013377924, 'number': 797}, 'P': {'precision': 0.050515463917525774, 'recall': 0.058682634730538925, 'f1': 0.05429362880886427, 'number': 835}, 'overall_precision': 0.039444850255661065, 'overall_recall': 0.03308823529411765, 'overall_f1': 0.035988003998667105, 'overall_accuracy': 0.5143524622394929}
			------------EPOCH 2---------------
Loss:  tensor(2268.5339, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1250.0936, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1936.2141, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1487.0127, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1045.4186, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1057.5824, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2578.3193, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1972.3704, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1744.1495, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(507.9106, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1043.1871, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2020.8658, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1147.7834, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2358.3489, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2636.0249, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1970.5847, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1605.5908, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1216.3734, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2273.2402, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(774.2963, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1673.0093, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1366.0980, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1489.8157, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1980.7067, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2134.5374, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1998.8984, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1814.1405, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1463.0874, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1203.1838, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1644.7587, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1329.9786, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1372.4241, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1886.3840, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1194.9200, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1093.9758, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.044919786096256686, 'recall': 0.05269761606022585, 'f1': 0.04849884526558891, 'number': 797}, 'P': {'precision': 0.0658578856152513, 'recall': 0.045508982035928146, 'f1': 0.05382436260623229, 'number': 835}, 'overall_precision': 0.05291005291005291, 'overall_recall': 0.049019607843137254, 'overall_f1': 0.05089058524173028, 'overall_accuracy': 0.5114611523604654}
			------------EPOCH 3---------------
Loss:  tensor(2036.4001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1118.0116, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1859.3065, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1361.8582, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(914.3506, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(891.8578, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2461.8689, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1746.5276, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1489.4849, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(459.2708, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(916.3289, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1722.6528, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1009.7670, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2062.0718, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2377.9792, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1691.0222, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1388.2599, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1047.8594, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2081.9775, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(671.4542, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1449.7166, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1164.1426, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1281.5784, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1967.7838, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1999.0820, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1884.2357, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1636.5081, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1269.8551, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(991.3024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1380.1918, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1125.3169, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1097.5562, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1511.1700, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1036.0326, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(961.2752, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.053203040173724216, 'recall': 0.06148055207026349, 'f1': 0.0570430733410943, 'number': 797}, 'P': {'precision': 0.059001512859304085, 'recall': 0.046706586826347304, 'f1': 0.05213903743315508, 'number': 835}, 'overall_precision': 0.05562579013906448, 'overall_recall': 0.05392156862745098, 'overall_f1': 0.054760423148724334, 'overall_accuracy': 0.5353780676797817}
			------------EPOCH 4---------------
Loss:  tensor(1828.8066, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(996.0239, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1588.3367, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1187.6779, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(834.5299, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(799.1052, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2380.4587, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1647.9050, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1386.1189, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(419.4634, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(845.9037, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1577.7511, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(900.2136, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1809.5348, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2094.7783, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1479.7159, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1084.4945, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(904.6764, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1813.0879, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(592.4349, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1162.5280, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(746.7836, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(852.7302, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1290.6956, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1651.4297, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1550.8284, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1441.0085, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1077.5991, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(900.6896, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1217.5396, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1019.3690, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(921.0245, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1374.1855, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(950.7151, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(880.0687, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.11526479750778816, 'recall': 0.09284818067754078, 'f1': 0.10284920083391245, 'number': 797}, 'P': {'precision': 0.20157068062827224, 'recall': 0.27664670658682633, 'f1': 0.2332155477031802, 'number': 835}, 'overall_precision': 0.17058165548098433, 'overall_recall': 0.1868872549019608, 'overall_f1': 0.1783625730994152, 'overall_accuracy': 0.5670205629958597}
			------------EPOCH 5---------------
Loss:  tensor(1761.8750, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(881.0385, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1251.1010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1081.3982, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(684.1826, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(676.1515, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1914.1504, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1316.4395, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1130.3278, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(316.5791, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(668.9637, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1298.6311, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(752.7544, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1487.1449, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1825.1273, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1250.1681, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(938.7377, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(806.5413, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1590.1848, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(535.6136, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1032.0840, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(536.9257, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(654.4901, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(990.7529, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1297.7583, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1329.6392, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1329.3535, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(853.0655, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(707.1559, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1046.5027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(741.3637, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(787.8244, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1031.4126, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(769.2853, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(749.1335, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.13237410071942446, 'recall': 0.11543287327478043, 'f1': 0.12332439678284182, 'number': 797}, 'P': {'precision': 0.16216216216216217, 'recall': 0.22994011976047904, 'f1': 0.19019316493313523, 'number': 835}, 'overall_precision': 0.15114422565194252, 'overall_recall': 0.17401960784313725, 'overall_f1': 0.16177727143264026, 'overall_accuracy': 0.5825179839474476}
			------------EPOCH 6---------------
Loss:  tensor(1392.7356, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(689.3019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1038.3777, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(881.2674, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(585.5232, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(566.7745, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1689.3060, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1127.9756, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1020.5381, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(246.0533, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(555.6095, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1164.4456, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(691.5596, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1368.0668, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1692.8367, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1007.3060, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(744.6389, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(580.7369, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1402.7383, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(430.5050, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(790.1239, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(391.4481, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(498.8414, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(705.7129, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(952.9774, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1064.2687, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1023.3531, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(651.7676, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(588.1483, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(935.4153, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(594.8319, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(660.1038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(827.7306, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(635.5151, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(696.8611, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.1681877444589309, 'recall': 0.1618569636135508, 'f1': 0.1649616368286445, 'number': 797}, 'P': {'precision': 0.08980827447023208, 'recall': 0.10658682634730539, 'f1': 0.09748083242059144, 'number': 835}, 'overall_precision': 0.12400455062571103, 'overall_recall': 0.13357843137254902, 'overall_f1': 0.12861356932153392, 'overall_accuracy': 0.5892952143038882}
			------------EPOCH 7---------------
Loss:  tensor(1207.0884, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(569.7695, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(869.2085, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(726.8679, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(383.0560, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(453.1910, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1395.2786, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(945.4902, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(807.4980, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(196.3266, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(441.5093, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(860.0621, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(572.1119, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1089.2279, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1503.4731, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(814.2734, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(670.4368, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(483.3651, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1256.1506, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(364.5604, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(644.5438, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(313.1157, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(365.9119, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(538.0029, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(918.1882, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(838.0670, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(896.5735, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(563.0317, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(462.6306, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(711.6920, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(458.1123, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(524.5692, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(635.7759, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(502.6415, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(564.7891, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.18195488721804512, 'recall': 0.15181932245922208, 'f1': 0.1655266757865937, 'number': 797}, 'P': {'precision': 0.07921714818266543, 'recall': 0.10179640718562874, 'f1': 0.08909853249475891, 'number': 835}, 'overall_precision': 0.11852704257767549, 'overall_recall': 0.12622549019607843, 'overall_f1': 0.12225519287833828, 'overall_accuracy': 0.5985011449587121}
			------------EPOCH 8---------------
Loss:  tensor(934.0855, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(434.3855, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(679.7206, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(628.2340, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(331.6069, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(398.1376, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1380.6182, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(833.2554, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(733.5171, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(179.7336, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(423.9803, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(808.3965, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(479.5391, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(923.6245, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1180.1241, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(672.1576, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(514.4766, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(338.0288, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(944.7011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(267.3635, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(516.7138, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(221.4987, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(278.3515, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(395.5716, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(700.5960, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(666.1799, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(885.2975, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(435.7364, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(396.6294, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(604.6635, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(391.1340, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(465.3183, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(634.2379, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(490.1853, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(519.9311, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.21337946943483277, 'recall': 0.23212045169385195, 'f1': 0.22235576923076925, 'number': 797}, 'P': {'precision': 0.16445623342175067, 'recall': 0.22275449101796407, 'f1': 0.18921668362156663, 'number': 835}, 'overall_precision': 0.1856856856856857, 'overall_recall': 0.22732843137254902, 'overall_f1': 0.20440771349862258, 'overall_accuracy': 0.5978997525038744}
			------------EPOCH 9---------------
Loss:  tensor(1033.7375, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(418.1037, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(611.2377, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(572.5654, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(234.4984, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(278.9543, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1232.2583, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(557.7851, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(544.9254, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(144.7111, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(323.7952, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(641.9828, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(463.6573, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(876.8773, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1069.8815, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(695.4049, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(565.6519, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(390.2430, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(933.6334, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(265.6749, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(513.7985, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(169.4274, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(229.9276, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(318.4085, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(497.2021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(476.3344, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(703.3672, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(334.6198, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(298.1281, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(478.8880, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(270.9454, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(328.0267, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(460.7885, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(363.2145, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(485.9003, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3115468409586057, 'recall': 0.1794228356336261, 'f1': 0.22770700636942676, 'number': 797}, 'P': {'precision': 0.17950391644908617, 'recall': 0.32934131736526945, 'f1': 0.23236163920574565, 'number': 835}, 'overall_precision': 0.20994475138121546, 'overall_recall': 0.25612745098039214, 'overall_f1': 0.23074799889594255, 'overall_accuracy': 0.6040755904054773}
			------------EPOCH 10---------------
Loss:  tensor(825.3116, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(360.4224, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(386.2717, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(467.2461, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(203.4849, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(298.7661, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(933.7438, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(421.2264, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(615.4556, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(132.7600, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(288.9047, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(682.8958, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(384.8533, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(782.5805, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(973.4800, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(529.7678, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(415.8914, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(206.0852, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(627.5585, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(155.7850, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(350.8767, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(117.6131, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(159.2220, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(196.9736, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(319.9825, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(307.6412, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(423.2108, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(233.3194, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(207.4660, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(412.2657, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(183.7143, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(276.2606, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(422.6371, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(251.0180, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(329.9489, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.2099236641221374, 'recall': 0.13801756587202008, 'f1': 0.1665404996214989, 'number': 797}, 'P': {'precision': 0.08851499634235552, 'recall': 0.14491017964071856, 'f1': 0.10990009082652133, 'number': 835}, 'overall_precision': 0.12215758857747223, 'overall_recall': 0.14154411764705882, 'overall_f1': 0.13113823445926767, 'overall_accuracy': 0.6168898757893276}
			------------EPOCH 11---------------
Loss:  tensor(570.5469, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(233.5914, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(364.7634, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(337.7106, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(139.9155, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(186.5997, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(981.4122, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(334.8371, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(451.5611, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(112.4623, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(230.7004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(487.0468, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(292.2808, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(569.9207, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(713.8014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(364.7771, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(301.3199, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(165.8045, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(457.4160, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(120.9391, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(353.8894, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(95.8586, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(143.7521, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(176.2159, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(328.3154, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(291.3472, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(535.4024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(224.5672, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(188.6387, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(297.0115, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(167.6238, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(219.8520, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(313.4366, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(188.5862, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(270.5524, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.24953445065176907, 'recall': 0.16813048933500627, 'f1': 0.20089955022488756, 'number': 797}, 'P': {'precision': 0.1696927374301676, 'recall': 0.29101796407185626, 'f1': 0.21438023820026467, 'number': 835}, 'overall_precision': 0.191467750126968, 'overall_recall': 0.23100490196078433, 'overall_f1': 0.2093862815884477, 'overall_accuracy': 0.6183933569264219}
			------------EPOCH 12---------------
Loss:  tensor(471.6002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(179.1721, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(198.1063, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(308.6658, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(126.2673, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(253.7866, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(673.0709, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(258.1551, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(396.6609, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(112.4923, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(319.5764, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(543.6896, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(304.9084, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(609.0640, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1033.7628, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(560.3284, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(377.1758, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(169.4488, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(424.9093, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(127.3517, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(289.3766, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(79.9817, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(98.6788, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(124.6349, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(528.7012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(392.8240, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(573.2679, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(240.7091, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(276.9272, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(482.7635, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(299.1516, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(324.5002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(421.7798, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(271.2309, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(283.2267, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.21029668411867364, 'recall': 0.30238393977415307, 'f1': 0.2480699948533196, 'number': 797}, 'P': {'precision': 0.13038793103448276, 'recall': 0.14491017964071856, 'f1': 0.13726602382302894, 'number': 835}, 'overall_precision': 0.17454194792671165, 'overall_recall': 0.22181372549019607, 'overall_f1': 0.19535887749595252, 'overall_accuracy': 0.6072213355538593}
			------------EPOCH 13---------------
Loss:  tensor(481.3492, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(179.7436, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(332.9242, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(292.1840, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(80.0287, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(149.6131, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(685.5539, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(198.9888, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(281.2176, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(83.0854, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(139.2053, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(283.9459, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(204.4986, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(372.6189, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(728.0984, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(361.4102, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(561.0849, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(447.7602, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(803.0902, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(347.4036, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(446.2498, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(154.8084, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(341.5022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(558.2837, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(394.2794, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(448.0818, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(788.7704, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(249.7506, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(186.2363, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(265.7995, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(143.2173, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(146.0770, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(273.8983, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(138.0111, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(192.8367, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.21844660194174756, 'recall': 0.33877038895859474, 'f1': 0.26561731431382196, 'number': 797}, 'P': {'precision': 0.1484848484848485, 'recall': 0.11736526946107785, 'f1': 0.1311036789297659, 'number': 835}, 'overall_precision': 0.1940928270042194, 'overall_recall': 0.22549019607843138, 'overall_f1': 0.20861678004535145, 'overall_accuracy': 0.5886244304119538}
			------------EPOCH 14---------------
Loss:  tensor(368.0488, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(167.9866, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(247.4570, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(363.9570, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(189.0265, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(297.0101, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1032.5847, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(539.0967, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(597.1302, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(157.3504, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(305.3962, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(739.2161, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(337.4177, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(707.7457, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(888.7335, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(571.2354, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(396.2609, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(217.3289, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(434.1874, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(106.6501, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(231.3007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(66.8767, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(79.4463, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(107.5208, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(155.2270, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(155.6273, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(382.0881, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(155.1687, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(194.0917, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(287.3633, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(158.8344, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(148.4502, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(301.3116, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(184.4479, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(281.0846, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.2661498708010336, 'recall': 0.12923462986198245, 'f1': 0.1739864864864865, 'number': 797}, 'P': {'precision': 0.1557759626604434, 'recall': 0.3197604790419162, 'f1': 0.2094939191839937, 'number': 835}, 'overall_precision': 0.17610661589719181, 'overall_recall': 0.2267156862745098, 'overall_f1': 0.1982319849986606, 'overall_accuracy': 0.6071750745957949}
			------------EPOCH 15---------------
Loss:  tensor(478.4745, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(168.3578, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(308.6974, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(381.1951, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(111.6159, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(132.6383, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(867.5999, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(253.1915, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(396.4299, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(67.5487, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(126.9581, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(341.4174, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(176.3662, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(309.1471, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(434.1616, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(236.1592, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(185.2544, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(76.6520, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(239.6583, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.3358, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(222.1245, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(59.1164, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.5669, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(87.8869, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(155.3931, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(96.7379, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(224.2361, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(91.0148, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(82.0073, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(137.2227, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(79.2280, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(90.5978, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(180.6588, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(102.2975, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(135.5443, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.19900990099009902, 'recall': 0.2521957340025094, 'f1': 0.22246817930271168, 'number': 797}, 'P': {'precision': 0.12532865907099036, 'recall': 0.17125748502994012, 'f1': 0.14473684210526316, 'number': 835}, 'overall_precision': 0.1599256159925616, 'overall_recall': 0.2107843137254902, 'overall_f1': 0.18186624372191385, 'overall_accuracy': 0.608655425253857}
			------------EPOCH 16---------------
Loss:  tensor(268.4999, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(100.8007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(153.4180, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(156.4712, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(61.0165, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(85.5381, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(553.5948, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(129.4458, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(273.6848, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(61.3113, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(93.9328, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(242.9762, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(143.3994, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(269.1025, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(404.9993, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(210.8207, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(219.3760, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(68.2037, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(170.8837, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.8816, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(152.5129, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.7840, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.5278, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.9000, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(82.1541, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.7707, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(178.7900, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(76.1503, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(77.5129, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(122.2231, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(61.9045, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(69.6825, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(116.8554, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(66.2087, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(94.8412, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.24123422159887797, 'recall': 0.21580928481806774, 'f1': 0.22781456953642384, 'number': 797}, 'P': {'precision': 0.1634689178818112, 'recall': 0.25508982035928146, 'f1': 0.1992516370439663, 'number': 835}, 'overall_precision': 0.1909722222222222, 'overall_recall': 0.23590686274509803, 'overall_f1': 0.21107456140350878, 'overall_accuracy': 0.6226724955473828}
			------------EPOCH 17---------------
Loss:  tensor(190.3767, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(67.0467, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(84.2876, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(178.9341, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.7510, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.7828, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(514.6868, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(82.1980, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(186.5922, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.5248, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(65.3447, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(182.0070, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(116.0054, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(193.4348, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(285.4131, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(139.3090, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(130.1807, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.1323, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(132.9257, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.9394, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(97.1527, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.4424, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.1165, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.2178, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.3022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.8881, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(162.2207, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.9193, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.3883, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(81.9256, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.5991, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.5704, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(102.2974, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.7911, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(81.0640, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.25910364145658266, 'recall': 0.23212045169385195, 'f1': 0.24487094639311718, 'number': 797}, 'P': {'precision': 0.1488469601677149, 'recall': 0.25508982035928146, 'f1': 0.18799646954986765, 'number': 835}, 'overall_precision': 0.18554778554778556, 'overall_recall': 0.24387254901960784, 'overall_f1': 0.21074927190892245, 'overall_accuracy': 0.6178613559086809}
			------------EPOCH 18---------------
Loss:  tensor(152.6238, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.4414, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(72.4377, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(123.9037, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.2146, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.3394, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(371.0453, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(59.0842, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(163.0032, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.5369, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.4591, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(130.0969, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(86.7349, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(141.6418, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(251.1370, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(111.3198, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(109.2194, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.0152, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(109.9719, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.0428, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(72.6281, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.0419, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.6145, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.5271, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.1462, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.2063, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(148.7575, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.2712, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.7206, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(62.9636, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.5168, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.1715, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(79.8857, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.8752, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.0261, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.27015558698727016, 'recall': 0.2396486825595985, 'f1': 0.25398936170212766, 'number': 797}, 'P': {'precision': 0.15378670788253476, 'recall': 0.23832335329341317, 'f1': 0.18694222639736965, 'number': 835}, 'overall_precision': 0.19490254872563717, 'overall_recall': 0.23897058823529413, 'overall_f1': 0.21469859620148635, 'overall_accuracy': 0.6241528462054449}
			------------EPOCH 19---------------
Loss:  tensor(148.5273, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.6758, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.1593, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(104.6699, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.4309, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.1609, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(324.0911, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.2869, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(139.0736, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.0754, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.3316, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(94.2814, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.7583, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(92.3983, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(212.7555, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(80.5445, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(94.7726, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.2487, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(82.6448, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.2198, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.8961, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.0370, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.6829, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.6732, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.6736, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.6686, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(131.4682, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.8739, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.0751, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.3048, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.3075, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.3328, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(61.4044, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.3648, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.4135, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.2379778051787916, 'recall': 0.24215809284818068, 'f1': 0.24004975124378106, 'number': 797}, 'P': {'precision': 0.1446776611694153, 'recall': 0.2311377245508982, 'f1': 0.17796219455970494, 'number': 835}, 'overall_precision': 0.17995337995337995, 'overall_recall': 0.23651960784313725, 'overall_f1': 0.20439502250463332, 'overall_accuracy': 0.6213540582425462}
			------------EPOCH 20---------------
Loss:  tensor(136.5907, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.1528, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.9346, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(79.5482, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.9047, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.6068, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(278.9846, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(67.6144, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(137.3359, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.9311, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.6450, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(65.6244, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.5586, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(69.3687, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(190.3322, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.7284, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(79.6888, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.6642, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(65.1544, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.9179, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.6251, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.3143, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.9579, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.4704, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.4119, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.6554, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(124.2064, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.1503, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.5931, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.1294, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.6866, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.3314, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.4314, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.7588, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.5485, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.24907063197026022, 'recall': 0.2521957340025094, 'f1': 0.25062344139650866, 'number': 797}, 'P': {'precision': 0.1702127659574468, 'recall': 0.25868263473053893, 'f1': 0.20532319391634982, 'number': 835}, 'overall_precision': 0.20086705202312138, 'overall_recall': 0.2555147058823529, 'overall_f1': 0.22491909385113268, 'overall_accuracy': 0.6223949297989961}
			------------EPOCH 21---------------
Loss:  tensor(114.0074, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.0826, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.7961, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(71.2327, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.8249, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.1241, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(252.4408, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.9389, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(124.0366, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.5497, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.0689, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.1190, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.2144, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.2621, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(165.4979, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.0989, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(68.0869, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.5509, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.5569, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.8103, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.3241, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.7175, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.4625, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.7793, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.5331, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.0354, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(119.4875, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.0731, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.6598, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.1313, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.9743, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.1599, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.4946, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.6628, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.6832, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.2402022756005057, 'recall': 0.2383939774153074, 'f1': 0.23929471032745592, 'number': 797}, 'P': {'precision': 0.16409861325115563, 'recall': 0.25508982035928146, 'f1': 0.19971870604781997, 'number': 835}, 'overall_precision': 0.192915270464337, 'overall_recall': 0.24693627450980393, 'overall_f1': 0.21660843859177645, 'overall_accuracy': 0.6243147595586704}
			------------EPOCH 22---------------
Loss:  tensor(104.1889, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.0180, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.0042, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(66.2677, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.3428, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.5607, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(225.4330, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.8154, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(108.2326, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.5212, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.2763, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.2666, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.3903, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.4074, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(142.9240, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.0635, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(62.7551, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.7178, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.7512, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.9019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.5633, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.7842, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.0144, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.5188, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.4098, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.3810, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(114.6236, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.5771, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.5769, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.6083, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.4968, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.1605, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.6690, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.2736, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.0633, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.2455470737913486, 'recall': 0.24215809284818068, 'f1': 0.24384080859128238, 'number': 797}, 'P': {'precision': 0.161698956780924, 'recall': 0.25988023952095807, 'f1': 0.19935691318327972, 'number': 835}, 'overall_precision': 0.19266917293233082, 'overall_recall': 0.2512254901960784, 'overall_f1': 0.2180851063829787, 'overall_accuracy': 0.623320148960285}
			------------EPOCH 23---------------
Loss:  tensor(104.0510, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.4928, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.5789, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.3894, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.2055, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.9877, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(215.0361, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.7806, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(98.4478, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.2642, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.8131, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.4304, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.0164, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.8330, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(138.1108, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.9105, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.1746, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.1125, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.8802, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.8077, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.7893, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.5444, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.9396, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.3651, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.3188, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.1021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(111.2973, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.4042, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.2432, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.4913, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.3672, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.1964, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.8902, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.0911, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.1126, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.24333719582850522, 'recall': 0.26348808030112925, 'f1': 0.2530120481927711, 'number': 797}, 'P': {'precision': 0.18174665617623917, 'recall': 0.27664670658682633, 'f1': 0.21937321937321932, 'number': 835}, 'overall_precision': 0.20665417057169636, 'overall_recall': 0.2702205882352941, 'overall_f1': 0.2342007434944238, 'overall_accuracy': 0.6194573589619041}
			------------EPOCH 24---------------
Loss:  tensor(91.3684, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.0243, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.1337, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.6548, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.0614, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.8038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(197.8392, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.7230, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(94.8574, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.0860, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.2034, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.0562, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.4002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.2581, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(131.7164, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.5145, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.2630, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.8372, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.7958, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.3898, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.4198, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.6719, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.7348, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.4097, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.4638, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.9115, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(114.2475, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.2548, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.2139, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.4719, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.1601, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.6067, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.2349, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.3881, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.9956, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.24081632653061225, 'recall': 0.22208281053952322, 'f1': 0.2310704960835509, 'number': 797}, 'P': {'precision': 0.14697406340057637, 'recall': 0.24431137724550897, 'f1': 0.18353576248313092, 'number': 835}, 'overall_precision': 0.17946302402260952, 'overall_recall': 0.23345588235294118, 'overall_f1': 0.2029294274300932, 'overall_accuracy': 0.6299123354844679}
			------------EPOCH 25---------------
Loss:  tensor(94.2837, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.6576, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.7740, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.4728, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.1962, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.9190, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(183.5986, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.6260, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(100.3844, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.0997, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.9282, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.3034, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.0986, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.0075, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(124.3876, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.7271, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.6335, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.6657, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.8501, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.7788, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.4573, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.7557, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.3698, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.5115, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.8719, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.2057, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(115.1832, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.9506, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.3964, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.6781, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.4445, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.4792, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.5086, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.4853, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.9751, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.22630992196209587, 'recall': 0.2547051442910916, 'f1': 0.2396694214876033, 'number': 797}, 'P': {'precision': 0.17592592592592593, 'recall': 0.27305389221556886, 'f1': 0.21398404504927263, 'number': 835}, 'overall_precision': 0.19653442772457821, 'overall_recall': 0.26409313725490197, 'overall_f1': 0.225359477124183, 'overall_accuracy': 0.6076839451345037}
			------------EPOCH 26---------------
Loss:  tensor(105.3628, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.2173, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.6439, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.2119, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.8158, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.2210, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(176.9865, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(115.8398, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(125.0124, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.5045, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.3070, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.2615, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.8364, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.5535, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(113.8618, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.2195, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.3438, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.0738, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.0237, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.2235, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.6221, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.9122, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.0789, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.7641, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.4635, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.3141, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(102.2524, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.2971, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.9717, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.4451, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.8628, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.2021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.0328, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.8966, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.6434, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.23684210526315788, 'recall': 0.2258469259723965, 'f1': 0.23121387283236994, 'number': 797}, 'P': {'precision': 0.1469559132260322, 'recall': 0.25149700598802394, 'f1': 0.18551236749116606, 'number': 835}, 'overall_precision': 0.17816354499771586, 'overall_recall': 0.23897058823529413, 'overall_f1': 0.204135043182413, 'overall_accuracy': 0.6247773691393149}
			------------EPOCH 27---------------
Loss:  tensor(108.5740, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.8152, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.4972, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.3080, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.6580, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.3894, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(161.1436, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.4378, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(109.7323, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.7431, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.5914, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.6399, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.4342, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.0200, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(95.3349, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.7614, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.3771, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.2791, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.6894, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.1066, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.9646, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.3202, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.5133, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.9134, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.9916, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.8920, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(110.1897, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.3748, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.3396, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.9461, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.1011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.9147, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.5278, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.8778, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.6804, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.2294478527607362, 'recall': 0.23462986198243413, 'f1': 0.23200992555831265, 'number': 797}, 'P': {'precision': 0.1751497005988024, 'recall': 0.28023952095808385, 'f1': 0.2155688622754491, 'number': 835}, 'overall_precision': 0.19572291957229196, 'overall_recall': 0.25796568627450983, 'overall_f1': 0.2225746761829236, 'overall_accuracy': 0.6165891795619087}
			------------EPOCH 28---------------
Loss:  tensor(101.4547, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.4181, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.9805, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.3885, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.5781, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.9847, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(176.0801, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.8316, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(103.2399, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.3623, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.3169, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.4165, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.6125, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(64.3101, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(108.9979, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.4075, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.3020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.3128, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.0058, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.1178, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.0464, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.9804, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.0565, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.3089, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.2325, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.9629, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(104.3695, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.9091, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.3160, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.2784, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.5830, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.2010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.0731, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.0265, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.8223, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.23122765196662692, 'recall': 0.24341279799247176, 'f1': 0.23716381418092908, 'number': 797}, 'P': {'precision': 0.17153558052434456, 'recall': 0.274251497005988, 'f1': 0.21105990783410136, 'number': 835}, 'overall_precision': 0.19457221711131556, 'overall_recall': 0.25919117647058826, 'overall_f1': 0.22228060956384657, 'overall_accuracy': 0.6175837901602942}
			------------EPOCH 29---------------
Loss:  tensor(70.0350, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.9906, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.3719, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.7133, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.3903, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.8016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(142.8485, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.9233, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(93.2810, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.9057, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.7668, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.3959, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.9048, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.4201, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(91.3834, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.4828, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.7960, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.1380, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.7375, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.4075, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.5851, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.2604, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.5776, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.2962, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.6648, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.7126, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(85.9438, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.1150, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.6562, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.9710, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.5079, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.3020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.5136, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.2575, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.0141, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.2403282532239156, 'recall': 0.2572145545796738, 'f1': 0.2484848484848485, 'number': 797}, 'P': {'precision': 0.17228177641653905, 'recall': 0.2694610778443114, 'f1': 0.21018215787015412, 'number': 835}, 'overall_precision': 0.19916628068550254, 'overall_recall': 0.26348039215686275, 'overall_f1': 0.22685307306779215, 'overall_accuracy': 0.621862928781255}
			------------EPOCH 30---------------
Loss:  tensor(60.2261, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.5496, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.3592, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.3560, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.9753, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.9594, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(120.1992, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(74.8719, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(100.8643, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5671, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.1155, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.8131, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.7961, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(64.6941, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(93.2643, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.9226, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.2600, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.6656, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.3997, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2608, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.8554, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.9970, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.1068, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.3274, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.5930, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.6864, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(77.4300, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.3126, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.4837, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.9661, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.3484, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.7226, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.8161, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.0769, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.5028, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.22511848341232227, 'recall': 0.2383939774153074, 'f1': 0.2315661182205972, 'number': 797}, 'P': {'precision': 0.1339820359281437, 'recall': 0.21437125748502994, 'f1': 0.16490096729617687, 'number': 835}, 'overall_precision': 0.16926605504587156, 'overall_recall': 0.2261029411764706, 'overall_f1': 0.19359916054564535, 'overall_accuracy': 0.6217010154280295}


		-------------RUN 3-----------
			------------EPOCH 1---------------
Loss:  tensor(2426.8481, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2410.8779, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1190.2474, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1540.2726, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1032.3534, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(713.5636, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1882.7910, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2971.6631, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1021.3881, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2105.6038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2096.9863, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1524.7812, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2516.2559, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2232.8857, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2793.2754, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1919.3647, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1887.5862, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2225.6826, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1607.5354, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2000.7693, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1403.7740, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1304.2654, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1453.2913, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(741.9692, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2341.5122, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2295.6558, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2126.8989, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1165.1979, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1759.1832, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1918.3068, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2617.9146, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2513.9597, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3135.5752, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2545.6797, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1938.6980, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'number': 720}, 'P': {'precision': 0.017543859649122806, 'recall': 0.009535160905840286, 'f1': 0.012355212355212355, 'number': 839}, 'overall_precision': 0.013468013468013467, 'overall_recall': 0.005131494547787043, 'overall_f1': 0.007431490942870413, 'overall_accuracy': 0.45402503622413837}
			------------EPOCH 2---------------
Loss:  tensor(1709.1937, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1603.0281, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(759.3474, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(924.0273, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(781.6545, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(531.1200, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1483.4574, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2385.5579, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(778.8196, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1681.0198, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1630.5293, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1246.5583, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2176.1621, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2091.5010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2457.7659, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1620.3735, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1799.2224, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1887.5420, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1357.8865, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1721.2936, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1242.9385, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1136.3756, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1304.3627, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(631.3083, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2078.0210, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2069.1792, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2038.9670, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(968.4276, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1385.2517, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1791.9652, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2008.0962, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2095.2844, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2871.9980, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2279.5405, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1745.6100, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.0136986301369863, 'recall': 0.022222222222222223, 'f1': 0.016949152542372878, 'number': 720}, 'P': {'precision': 0.03995560488346282, 'recall': 0.04290822407628129, 'f1': 0.041379310344827586, 'number': 839}, 'overall_precision': 0.02513291445142581, 'overall_recall': 0.03335471456061578, 'overall_f1': 0.028665931642778388, 'overall_accuracy': 0.5358322050404998}
			------------EPOCH 3---------------
Loss:  tensor(1551.9331, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1494.5249, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(691.8760, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(844.2445, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(725.8683, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(512.3918, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1387.8223, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2206.8672, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(700.4422, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1537.9396, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1431.1962, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1094.8317, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1911.8347, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1919.1609, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2145.1064, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1311.8480, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1717.8774, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1695.1580, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1140.4983, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1470.3607, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1105.7457, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1011.6459, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1173.8979, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(534.3120, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1856.7925, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1932.3862, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1955.5439, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(845.1189, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(990.2029, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1818.6543, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1581.1575, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1714.6647, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2777.0146, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2143.9092, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1726.5544, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.12615384615384614, 'recall': 0.05694444444444444, 'f1': 0.0784688995215311, 'number': 720}, 'P': {'precision': 0.05029838022165388, 'recall': 0.07032181168057211, 'f1': 0.05864811133200795, 'number': 839}, 'overall_precision': 0.06675567423230974, 'overall_recall': 0.06414368184733804, 'overall_f1': 0.06542361792607132, 'overall_accuracy': 0.5543362075108672}
			------------EPOCH 4---------------
Loss:  tensor(1397.2295, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1323.0322, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(579.1845, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(726.6460, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(633.4457, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(493.4232, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1233.7910, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2008.6511, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(629.9884, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1390.5398, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1290.0347, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1003.5140, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1684.9950, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1712.4846, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1735.7585, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1085.5482, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1614.5013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1529.2122, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(954.4490, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1331.6721, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(988.5456, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(912.5623, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1072.0745, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(469.4470, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1724.9407, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1807.8926, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1857.0724, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(787.0819, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(953.3747, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1724.6892, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1453.5100, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1555.2866, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2345.1619, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1829.4172, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1468.8674, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.068359375, 'recall': 0.04861111111111111, 'f1': 0.056818181818181816, 'number': 720}, 'P': {'precision': 0.1111111111111111, 'recall': 0.14541120381406436, 'f1': 0.12596799173980383, 'number': 839}, 'overall_precision': 0.09751552795031056, 'overall_recall': 0.10070558050032072, 'overall_f1': 0.09908488482171031, 'overall_accuracy': 0.5673056367134611}
			------------EPOCH 5---------------
Loss:  tensor(1267.5656, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1149.8529, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(481.7175, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(587.4957, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(540.4640, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(419.2583, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1128.3508, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1775.0808, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(554.7734, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1202.0925, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1137.2419, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(914.0188, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1559.6649, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1760.8267, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1609.8081, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1027.8962, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1292.1604, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1476.5977, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(992.0626, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1356.1173, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(895.6497, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(810.9131, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(951.6759, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(409.6581, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1539.2827, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1535.1711, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1482.0186, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(687.1559, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(855.4583, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1557.8613, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1331.4392, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1408.0765, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2169.3484, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1658.8594, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1286.3940, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.08442776735459662, 'recall': 0.125, 'f1': 0.10078387458006718, 'number': 720}, 'P': {'precision': 0.29619047619047617, 'recall': 0.3706793802145411, 'f1': 0.3292747485442033, 'number': 839}, 'overall_precision': 0.1895085066162571, 'overall_recall': 0.2572161642078255, 'overall_f1': 0.2182312925170068, 'overall_accuracy': 0.5788498539157699}
			------------EPOCH 6---------------
Loss:  tensor(1187.3276, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1103.2793, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(453.5361, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(526.5312, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(457.5417, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(319.5868, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(966.1235, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1609.9344, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(434.4563, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(970.7217, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(965.3052, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(828.6088, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1377.4608, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1345.9019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1468.6189, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(813.2849, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1122.6101, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1247.6538, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(810.2178, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1188.0671, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(792.4602, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(731.4796, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(861.7409, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(347.8403, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1417.6876, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1563.2600, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1343.8484, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(647.9379, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(620.1981, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1148.0016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1169.6228, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1422.7725, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1856.0178, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1440.6230, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1087.1042, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.09581497797356828, 'recall': 0.12083333333333333, 'f1': 0.10687960687960689, 'number': 720}, 'P': {'precision': 0.24842484248424843, 'recall': 0.3289630512514899, 'f1': 0.28307692307692306, 'number': 839}, 'overall_precision': 0.17979197622585438, 'overall_recall': 0.23284156510583706, 'overall_f1': 0.2029066517607602, 'overall_accuracy': 0.596379961519276}
			------------EPOCH 7---------------
Loss:  tensor(1004.1537, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(922.9629, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(356.8470, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(387.3568, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(370.7477, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(285.7449, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(858.3507, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1443.6650, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(365.0704, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(847.5511, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(863.6366, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(796.6636, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1360.4382, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1324.1316, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1812.7020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(760.9884, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1056.4626, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1030.6100, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(640.3955, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1089.2566, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(659.8250, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(617.4722, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(694.9655, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(298.4384, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1200.0835, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1266.4802, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1129.7688, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(503.7444, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(522.4419, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1020.9818, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(962.1606, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1083.5183, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1574.7219, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1229.0663, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(876.4558, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.07272727272727272, 'recall': 0.10555555555555556, 'f1': 0.08611898016997167, 'number': 720}, 'P': {'precision': 0.08961303462321792, 'recall': 0.10488676996424315, 'f1': 0.09665019220208675, 'number': 839}, 'overall_precision': 0.08090774543660582, 'overall_recall': 0.10519563822963438, 'overall_f1': 0.09146681539319576, 'overall_accuracy': 0.6064277061212855}
			------------EPOCH 8---------------
Loss:  tensor(861.9103, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(857.5099, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(288.7556, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(316.1662, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(288.5379, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(209.3676, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(801.0231, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1304.5422, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(298.1895, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(658.1161, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(692.9364, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(612.0574, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1045.5647, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1003.5657, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1139.3915, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(535.2123, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(814.2574, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(939.6777, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(550.4952, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(990.3952, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(587.9263, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(500.2307, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(614.3182, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(236.0349, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(983.8235, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(993.8099, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(826.7629, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(408.3111, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(395.2072, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(773.7457, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(775.9771, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(880.6516, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1319.6149, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(968.0367, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(684.2700, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.10922330097087378, 'recall': 0.1875, 'f1': 0.1380368098159509, 'number': 720}, 'P': {'precision': 0.24339839265212398, 'recall': 0.2526817640047676, 'f1': 0.247953216374269, 'number': 839}, 'overall_precision': 0.1646891314665401, 'overall_recall': 0.22257857601026299, 'overall_f1': 0.1893071467539553, 'overall_accuracy': 0.6160478871232096}
			------------EPOCH 9---------------
Loss:  tensor(721.6283, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(675.0861, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(240.6584, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(212.8065, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(206.6172, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(137.8190, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(587.3636, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1000.3911, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(222.6985, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(500.3277, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(532.9691, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(498.9790, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(829.3118, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(747.2919, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(996.3766, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(419.8195, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(598.5663, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(732.2297, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(431.7000, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(815.9047, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(482.9623, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(402.0826, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(498.2864, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(178.3425, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(787.2549, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(839.3633, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(676.4148, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(349.7263, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(293.5366, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(524.6157, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(658.8497, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(766.5433, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1013.3622, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(785.1971, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(497.3367, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.09543230016313213, 'recall': 0.1625, 'f1': 0.12024665981500514, 'number': 720}, 'P': {'precision': 0.24948665297741274, 'recall': 0.2896305125148987, 'f1': 0.2680639823496966, 'number': 839}, 'overall_precision': 0.16363636363636364, 'overall_recall': 0.23091725465041693, 'overall_f1': 0.19154030327214686, 'overall_accuracy': 0.6330079099266016}
			------------EPOCH 10---------------
Loss:  tensor(572.8461, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(506.3195, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(164.9569, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(151.0257, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(138.8941, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(98.2969, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(480.6426, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(772.6399, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(156.9630, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(350.7469, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(451.5424, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(408.1703, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(687.2584, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(588.1600, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(935.0677, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(291.8834, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(451.8036, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(546.3226, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(313.7101, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(666.8375, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(415.8627, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(305.2568, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(399.6038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(148.1898, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(554.2981, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(720.9397, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(529.3425, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(242.3181, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(223.6767, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(369.7827, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(485.7067, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(511.9136, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(880.8689, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(681.6245, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(421.9384, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.07914943886591849, 'recall': 0.18611111111111112, 'f1': 0.11106506423539161, 'number': 720}, 'P': {'precision': 0.1171003717472119, 'recall': 0.07508939213349225, 'f1': 0.09150326797385622, 'number': 839}, 'overall_precision': 0.08830121021963246, 'overall_recall': 0.12636305323925592, 'overall_f1': 0.10395778364116094, 'overall_accuracy': 0.5871398370507613}
			------------EPOCH 11---------------
Loss:  tensor(571.6921, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(530.3500, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(160.4731, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(151.9456, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(141.6213, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(88.1808, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(478.7300, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(878.2083, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(112.4008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(263.5729, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(429.1873, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(413.0771, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(840.1434, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(500.4401, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(741.6791, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(187.4110, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(437.1583, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(523.9999, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(325.1992, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(702.3330, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(383.6475, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(267.2749, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(337.0946, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(141.9663, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(403.9324, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(567.6146, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(415.5664, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(223.2433, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(177.0653, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(379.6171, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(413.9986, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(447.6454, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1038.6000, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(847.2780, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(493.8460, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.1371801850843767, 'recall': 0.35, 'f1': 0.19710598357450138, 'number': 720}, 'P': {'precision': 0.2109704641350211, 'recall': 0.11918951132300358, 'f1': 0.15232292460015232, 'number': 839}, 'overall_precision': 0.1523150151449589, 'overall_recall': 0.22578576010262988, 'overall_f1': 0.1819121447028424, 'overall_accuracy': 0.5648115157129623}
			------------EPOCH 12---------------
Loss:  tensor(479.9794, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(360.3294, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(137.9494, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(99.5248, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(113.9225, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(75.6639, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(343.9710, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(626.6625, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(104.7198, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(291.9210, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(404.0421, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(404.2939, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(633.1571, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(585.2465, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(731.9236, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(181.4787, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(316.2533, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(429.6145, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(268.9550, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(566.2321, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(376.4289, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(243.0780, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(365.0978, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(115.4837, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(771.8872, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1001.5321, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(756.4198, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(254.3957, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(164.6490, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(647.8434, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(575.0369, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(755.6946, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1371.8982, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1026.6514, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(544.4910, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.15164433617539586, 'recall': 0.3458333333333333, 'f1': 0.21083827265029634, 'number': 720}, 'P': {'precision': 0.20089955022488756, 'recall': 0.15971394517282478, 'f1': 0.17795484727755642, 'number': 839}, 'overall_precision': 0.16587267215244694, 'overall_recall': 0.2456703014753047, 'overall_f1': 0.19803516028955537, 'overall_accuracy': 0.5833155181833297}
			------------EPOCH 13---------------
Loss:  tensor(415.9523, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(296.7451, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(114.8004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(72.8988, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(158.7993, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(89.9616, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(404.7021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(644.7019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(329.1374, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(886.5930, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(826.6769, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(616.7540, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1129.8674, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1228.4536, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1381.7256, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(698.7990, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(705.3246, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(714.3704, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(511.5836, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(974.1166, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(274.3724, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(201.2377, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(262.2660, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(112.9674, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(419.9017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(497.3220, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(386.4797, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(199.1579, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(224.6823, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(569.9835, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(484.3781, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(590.2750, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1705.4514, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1272.2739, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1278.5774, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.17228464419475656, 'recall': 0.06388888888888888, 'f1': 0.09321175278622086, 'number': 720}, 'P': {'precision': 0.21159029649595687, 'recall': 0.3742550655542312, 'f1': 0.27034007748600947, 'number': 839}, 'overall_precision': 0.20559680182752713, 'overall_recall': 0.23091725465041693, 'overall_f1': 0.2175226586102719, 'overall_accuracy': 0.5866647663839997}
			------------EPOCH 14---------------
Loss:  tensor(1138.7706, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(743.6353, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(179.6235, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(326.7374, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(188.5177, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(178.6762, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1091.2759, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1510.9995, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(255.9758, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(559.9045, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(641.5425, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(532.6616, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(697.3924, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(573.9777, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(725.5973, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(337.0458, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(348.9594, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(492.4510, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(297.1282, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(546.8030, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(316.2855, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(237.1142, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(309.2149, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(127.5279, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(459.9471, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(625.2648, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(401.5349, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(262.4840, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(435.2846, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(339.2085, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(706.0015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(764.4197, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(868.0092, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(697.8930, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(440.2407, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.12251056125528063, 'recall': 0.28194444444444444, 'f1': 0.1708035338662179, 'number': 720}, 'P': {'precision': 0.18410041841004185, 'recall': 0.1573301549463647, 'f1': 0.16966580976863752, 'number': 839}, 'overall_precision': 0.14111204717775905, 'overall_recall': 0.21488133418858243, 'overall_f1': 0.1703534197813374, 'overall_accuracy': 0.5785410579823749}
			------------EPOCH 15---------------
Loss:  tensor(529.6819, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(376.7086, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(130.3481, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(120.6944, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(109.6633, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.7860, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(316.8566, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(531.6002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(101.6184, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(204.5293, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(261.4898, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(248.8037, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(469.1855, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(405., device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(625.1997, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(185.0251, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(280.1285, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(359.9453, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(218.6057, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(467.3477, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(254.7670, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(187.5869, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(220.3407, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(114.2533, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(254.8985, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(428.5692, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(260.3249, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(169.8675, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(135.4503, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(191.9749, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(300.1180, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(271.8322, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(568.8326, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(403.0790, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(242.0204, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.16441717791411042, 'recall': 0.18611111111111112, 'f1': 0.17459283387622151, 'number': 720}, 'P': {'precision': 0.16152597402597402, 'recall': 0.23718712753277713, 'f1': 0.19217769193626266, 'number': 839}, 'overall_precision': 0.1626770884220811, 'overall_recall': 0.21359846055163567, 'overall_f1': 0.1846921797004992, 'overall_accuracy': 0.6364284187272857}
			------------EPOCH 16---------------
Loss:  tensor(368.5448, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(257.2937, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(61.9713, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(70.6181, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(67.1215, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.2003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(296.4784, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(457.0172, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(65.7113, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(157.0784, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(231.0526, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(224.3024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(358.5085, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(322.4358, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(499.9780, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(121.3010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(180.5295, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(228.7495, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(144.7725, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(335.3608, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(183.7195, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(119.9980, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(155.5979, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(78.5380, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(166.8492, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(260.8498, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(187.6588, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(113.0566, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(86.6147, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(145.0402, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(233.7299, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(199.1611, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(457.0453, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(356.6388, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(179.0588, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.1935483870967742, 'recall': 0.26666666666666666, 'f1': 0.22429906542056074, 'number': 720}, 'P': {'precision': 0.22755555555555557, 'recall': 0.30512514898688914, 'f1': 0.26069246435845217, 'number': 839}, 'overall_precision': 0.21162021728861596, 'overall_recall': 0.2873636946760744, 'overall_f1': 0.24374319912948858, 'overall_accuracy': 0.6408228223948312}
			------------EPOCH 17---------------
Loss:  tensor(259.3679, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(165.7240, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.1656, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.2501, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.7282, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.2606, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(203.9189, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(327.4533, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.0480, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(91.5868, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(135.8811, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(149.3524, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(242.7969, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(224.9068, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(401.8637, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(90.7900, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(125.7104, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(189.8522, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(113.1750, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(282.8478, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(133.1007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(85.4161, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(103.6609, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.0227, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(124.7099, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(213.5681, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(143.9311, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(85.9370, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(74.2378, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(100.4897, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(176.3699, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(151.3239, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(380.9765, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(299.6366, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(140.9564, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.1720321931589537, 'recall': 0.2375, 'f1': 0.19953325554259038, 'number': 720}, 'P': {'precision': 0.18884892086330934, 'recall': 0.25029797377830754, 'f1': 0.21527421834956434, 'number': 839}, 'overall_precision': 0.18091168091168092, 'overall_recall': 0.24438742783835793, 'overall_f1': 0.20791268758526604, 'overall_accuracy': 0.6456447896624623}
			------------EPOCH 18---------------
Loss:  tensor(208.7696, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(117.9382, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.0853, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.3753, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.6941, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.6844, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(179.6586, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(271.2490, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.8599, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(67.0230, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(110.7052, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(125.1635, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(193.0459, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(172.2670, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(359.4576, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.0261, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(98.2878, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(115.3789, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(78.3645, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(211.5948, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(102.4370, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.3965, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(78.1149, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.0136, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(85.5548, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(172.0857, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(110.5511, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.8976, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.4399, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(77.1981, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(128.5794, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(106.8271, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(349.7737, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(271.5009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(121.7684, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.1659919028340081, 'recall': 0.22777777777777777, 'f1': 0.19203747072599534, 'number': 720}, 'P': {'precision': 0.19911894273127753, 'recall': 0.2693682955899881, 'f1': 0.22897669706180349, 'number': 839}, 'overall_precision': 0.18370230805463966, 'overall_recall': 0.2501603592046183, 'overall_f1': 0.2118413905486149, 'overall_accuracy': 0.652081997197083}
			------------EPOCH 19---------------
Loss:  tensor(176.3777, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(91.7568, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.4247, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.0277, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.1620, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.0215, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(151.9638, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(236.7533, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.2460, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.4063, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(85.7833, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(92.4396, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(146.3036, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(155.8658, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(357.6235, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.0212, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(76.5006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(85.0648, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.8564, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(181.0280, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(77.8510, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.1434, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.3893, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.5839, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.7629, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(162.1337, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(87.1593, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.6414, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.4736, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.3608, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(115.8211, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(86.7154, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(320.4196, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(240.0392, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(103.9875, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.18731417244796827, 'recall': 0.2625, 'f1': 0.2186234817813765, 'number': 720}, 'P': {'precision': 0.21739130434782608, 'recall': 0.29201430274135876, 'f1': 0.249237029501526, 'number': 839}, 'overall_precision': 0.20318352059925093, 'overall_recall': 0.2783835792174471, 'overall_f1': 0.23491204330175913, 'overall_accuracy': 0.6498491650633031}
			------------EPOCH 20---------------
Loss:  tensor(158.2692, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(72.9290, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.1953, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.1090, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.3174, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.8524, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(122.5277, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(206.8590, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.4983, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.7682, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.2287, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(68.8212, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(102.3450, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(146.4751, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(367.2087, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.3604, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.4203, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(64.0180, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.5907, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(151.8689, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.6004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.8139, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.2404, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.1638, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.2485, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(143.2269, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(66.5059, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.0850, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.3228, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.8661, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(108.2150, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(68.8876, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(288.2029, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(198.5838, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(82.5737, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.18575851393188855, 'recall': 0.25, 'f1': 0.21314387211367672, 'number': 720}, 'P': {'precision': 0.20496894409937888, 'recall': 0.27532777115613827, 'f1': 0.23499491353001017, 'number': 839}, 'overall_precision': 0.19608778625954199, 'overall_recall': 0.26363053239255935, 'overall_f1': 0.22489740082079346, 'overall_accuracy': 0.6535072091973682}
			------------EPOCH 21---------------
Loss:  tensor(144.1843, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.2273, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.6588, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.5605, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.7191, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.0376, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(96.1085, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(184.9522, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.7040, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.3817, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(59.9848, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.4800, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(88.7963, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(108.1780, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(288.7384, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.8877, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.5236, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.4995, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.3690, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(132.0717, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.4391, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.4056, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.7103, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.7450, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.7762, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(127.4202, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.6639, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.4838, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.2329, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.9654, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(101.9860, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.9408, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(264.4641, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(169.6928, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(69.9407, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.16363636363636364, 'recall': 0.225, 'f1': 0.1894736842105263, 'number': 720}, 'P': {'precision': 0.18448275862068966, 'recall': 0.2550655542312277, 'f1': 0.2141070535267634, 'number': 839}, 'overall_precision': 0.17488372093023255, 'overall_recall': 0.241180243745991, 'overall_f1': 0.20275006740361282, 'overall_accuracy': 0.6525095607971686}
			------------EPOCH 22---------------
Loss:  tensor(143.0618, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.8522, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.7002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.9478, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.1755, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.3093, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(75.6494, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(172.2022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.4800, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.9611, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(67.4268, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(69.6010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(97.2764, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(102.5965, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(266.3470, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.4517, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.6895, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.4026, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.1587, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(111.3057, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.3789, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.8911, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.3695, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.7234, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.8097, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(111.8642, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.3214, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.8509, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.2805, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.0715, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(76.8111, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.7414, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(257.6074, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(156.2817, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(62.2864, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.19159335288367546, 'recall': 0.2722222222222222, 'f1': 0.2248995983935743, 'number': 720}, 'P': {'precision': 0.22181818181818183, 'recall': 0.2908224076281287, 'f1': 0.2516761217122228, 'number': 839}, 'overall_precision': 0.20725388601036268, 'overall_recall': 0.28223220012828737, 'overall_f1': 0.23900054318305267, 'overall_accuracy': 0.646713698662676}
			------------EPOCH 23---------------
Loss:  tensor(129.1198, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.8195, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.5657, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.0042, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.1476, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.3098, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.9034, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(156.7689, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.9700, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.1287, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.1229, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.8157, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(66.2342, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(90.9856, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(250.8450, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.0590, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.0942, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.8180, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.8693, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(99.6947, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.4229, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.3278, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.7715, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.6725, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.4779, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(105.0271, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.9397, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.6538, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.2300, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.8745, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(80.0765, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.4855, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(257.5853, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(172.8122, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(70.9442, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.1921487603305785, 'recall': 0.25833333333333336, 'f1': 0.22037914691943128, 'number': 720}, 'P': {'precision': 0.20665499124343256, 'recall': 0.28128724672228844, 'f1': 0.23826350328117113, 'number': 839}, 'overall_precision': 0.2, 'overall_recall': 0.27068633739576653, 'overall_f1': 0.2300354319978196, 'overall_accuracy': 0.651511912396969}
			------------EPOCH 24---------------
Loss:  tensor(109.6502, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.2724, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.9565, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.7003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.4945, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.8571, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.2030, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(142.6877, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.4985, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.1435, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.5609, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.2238, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.4753, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(74.3732, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(263.7458, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.8117, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.4155, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.2685, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.7389, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(111.2110, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.3419, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.9027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.5928, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.3027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.8268, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(107.7673, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.8604, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.6630, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.7554, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.5630, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(68.8625, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.6462, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(349.0020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(344.6371, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(165.6965, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.17618586640851888, 'recall': 0.25277777777777777, 'f1': 0.2076440387906446, 'number': 720}, 'P': {'precision': 0.18287243532560213, 'recall': 0.24433849821215733, 'f1': 0.20918367346938774, 'number': 839}, 'overall_precision': 0.1796657381615599, 'overall_recall': 0.2482360487491982, 'overall_f1': 0.2084567734985187, 'overall_accuracy': 0.6471887693294378}
			------------EPOCH 25---------------
Loss:  tensor(102.5524, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.8542, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.1571, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.3845, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.5273, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5290, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.4437, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(150.6354, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.4176, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.1735, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.3431, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.4307, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(61.3116, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(80.4213, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(296.8580, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.6777, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.8219, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.6133, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.0313, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(133.7069, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.4598, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.2456, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.1685, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.9140, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.0645, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(132.8736, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.9582, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.9982, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.2422, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.5512, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(77.0634, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.7802, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(255.0996, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(174.7592, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(65.2383, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.19868995633187772, 'recall': 0.25277777777777777, 'f1': 0.22249388753056237, 'number': 720}, 'P': {'precision': 0.19656019656019655, 'recall': 0.2860548271752086, 'f1': 0.23300970873786409, 'number': 839}, 'overall_precision': 0.19747309312119793, 'overall_recall': 0.27068633739576653, 'overall_f1': 0.22835497835497834, 'overall_accuracy': 0.6426755979952018}
			------------EPOCH 26---------------
Loss:  tensor(110.0700, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.2052, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.6559, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.8228, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.6786, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.1485, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.5627, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(123.1260, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.1385, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.6445, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.1209, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.9497, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.4174, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(61.4708, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(277.8487, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.3820, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.8340, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.5482, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.8660, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(94.7199, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.5581, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.0165, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.1705, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.6874, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.7986, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(104.4215, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.0564, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.6519, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.2693, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.3402, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(108.2691, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.8073, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(209.8716, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(147.4890, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.4961, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.19500480307396734, 'recall': 0.28194444444444444, 'f1': 0.23055082339579783, 'number': 720}, 'P': {'precision': 0.21241258741258742, 'recall': 0.2896305125148987, 'f1': 0.24508320726172467, 'number': 839}, 'overall_precision': 0.20411899313501145, 'overall_recall': 0.28608082103912763, 'overall_f1': 0.23824786324786326, 'overall_accuracy': 0.639065060927813}
			------------EPOCH 27---------------
Loss:  tensor(104.7257, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.4881, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.5456, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.1685, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.8956, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.8852, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.4884, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(114.6662, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.0273, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.9036, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.0772, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.1258, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(71.4023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(67.6664, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(237.9117, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.9437, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.4373, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.8308, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.5971, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(85.3643, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.3529, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.4545, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.9907, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.4059, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.7004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(96.2380, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.6781, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.9646, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.8406, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.4889, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(83.1720, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.8357, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(201.8614, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(127.6142, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.6009, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.19017763845350052, 'recall': 0.25277777777777777, 'f1': 0.21705426356589147, 'number': 720}, 'P': {'precision': 0.17803660565723795, 'recall': 0.2550655542312277, 'f1': 0.20970112689857914, 'number': 839}, 'overall_precision': 0.18341824918943955, 'overall_recall': 0.25400898011545864, 'overall_f1': 0.21301775147928995, 'overall_accuracy': 0.644148317062163}
			------------EPOCH 28---------------
Loss:  tensor(97.5623, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.3731, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.6623, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.8180, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.1229, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.0505, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.9042, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(125.8593, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.0869, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.1220, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.4423, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.4161, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.6846, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(122.5556, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(215.6953, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.2809, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.4081, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.0884, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.3160, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(74.5610, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.7691, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.4602, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.9566, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.5601, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.2902, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(78.2907, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.1170, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.6336, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.3308, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.0255, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(67.3113, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.6581, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(183.7887, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(109.6988, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.1845, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.19523809523809524, 'recall': 0.2847222222222222, 'f1': 0.23163841807909605, 'number': 720}, 'P': {'precision': 0.20452640402347025, 'recall': 0.2908224076281287, 'f1': 0.24015748031496065, 'number': 839}, 'overall_precision': 0.20017833259028087, 'overall_recall': 0.28800513149454776, 'overall_f1': 0.23619147816938452, 'overall_accuracy': 0.6382811943276563}
			------------EPOCH 29---------------
Loss:  tensor(101.9044, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.3844, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.7415, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.6035, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.1791, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4226, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(68.3368, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(202.8057, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.3926, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.6230, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.5682, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.6173, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(73.3125, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(165.5804, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(223.6135, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.8991, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.2865, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.2458, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.3977, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(65.9837, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.4635, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.8267, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.0696, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.1102, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.3061, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(74.9497, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.5859, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.3523, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.7601, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.5050, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.4421, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.9711, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(165.7608, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(87.0675, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.7282, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.18716577540106952, 'recall': 0.24305555555555555, 'f1': 0.21148036253776437, 'number': 720}, 'P': {'precision': 0.1607717041800643, 'recall': 0.23837902264600716, 'f1': 0.19203072491598658, 'number': 839}, 'overall_precision': 0.1720972923359339, 'overall_recall': 0.24053880692751764, 'overall_f1': 0.20064205457463885, 'overall_accuracy': 0.6397301598612793}
			------------EPOCH 30---------------
Loss:  tensor(97.1607, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.4853, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.2447, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.8488, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.5364, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4595, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.5473, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(190.8983, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.0877, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.5334, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.5996, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.6094, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(71.7861, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(85.9797, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(210.6713, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.7884, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.2417, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.0999, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.1995, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.2257, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.3804, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.0263, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.3000, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.7775, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.6836, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(69.6124, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.0064, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.7451, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.7869, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.6290, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.3284, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.3544, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(146.3781, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(73.8866, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.1483, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.2030612244897959, 'recall': 0.2763888888888889, 'f1': 0.2341176470588235, 'number': 720}, 'P': {'precision': 0.20819397993311037, 'recall': 0.2967818831942789, 'f1': 0.2447174447174447, 'number': 839}, 'overall_precision': 0.20588235294117646, 'overall_recall': 0.2873636946760744, 'overall_f1': 0.23989290495314594, 'overall_accuracy': 0.6396588992612651}


		-------------RUN 4-----------
			------------EPOCH 1---------------
Loss:  tensor(2294.4004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2629.5129, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3399.1829, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2583.4241, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1455.4097, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1661.5034, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4928.8828, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1984.6738, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2419.8850, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2163.6406, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1523.6724, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1092.4458, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1133.0884, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1212.8981, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3035.7593, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2982.2012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1378.0962, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1226.4603, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1199.3911, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1462.2754, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2265.7029, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2681.3987, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2401.5872, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2553.2593, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1567.9604, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1647.6278, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1702.9178, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1101.7578, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(905.2467, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1638.7520, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2025.6970, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1625.2104, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1721.2334, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1740.8026, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2113.0273, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.014492753623188406, 'recall': 0.0012135922330097086, 'f1': 0.002239641657334826, 'number': 824}, 'P': {'precision': 0.041821561338289966, 'recall': 0.04643962848297214, 'f1': 0.044009779951100246, 'number': 969}, 'overall_precision': 0.04017467248908297, 'overall_recall': 0.025655326268823202, 'overall_f1': 0.031313818924438394, 'overall_accuracy': 0.507255139056832}
			------------EPOCH 2---------------
Loss:  tensor(1449.8796, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1717.2158, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2265.2937, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1693.2096, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1216.0098, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1319.2343, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4084.7734, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1523.5408, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2150.6458, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1928.6580, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1225.4341, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(892.1061, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(961.3610, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(977.2554, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2405.6375, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2342.6069, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1230.5358, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1086.8751, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1045.9961, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1360.4503, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2070.6621, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2441.6877, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2098.8076, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2265.6084, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1439.7017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1561.5261, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1573.2517, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(944.7075, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(690.2538, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1435.2778, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1660.6929, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1258.6111, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1364.9795, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1409.5479, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1907.8865, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.02834008097165992, 'recall': 0.00849514563106796, 'f1': 0.013071895424836602, 'number': 824}, 'P': {'precision': 0.08463726884779517, 'recall': 0.12280701754385964, 'f1': 0.10021052631578949, 'number': 969}, 'overall_precision': 0.07622504537205081, 'overall_recall': 0.07027328499721137, 'overall_f1': 0.07312826465467208, 'overall_accuracy': 0.5534954543418872}
			------------EPOCH 3---------------
Loss:  tensor(1227.8066, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1459.9978, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2117.6318, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1566.4709, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1030.8945, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1108.7167, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3652.0752, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1413.5808, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1957.6024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1701.7224, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1047.5367, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(773.8690, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(849.8588, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(888.1631, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2010.0837, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2107.8538, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1128.7233, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(958.1134, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(961.0130, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1263.3635, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1962.7791, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2118.3210, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1653.9050, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1945.9039, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1273.8281, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1265.9617, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1424.0955, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(867.4767, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(632.8356, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1244.3823, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1426.1788, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(938.6066, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1060.1816, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1147.3428, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1694.4856, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.0584958217270195, 'recall': 0.025485436893203883, 'f1': 0.03550295857988166, 'number': 824}, 'P': {'precision': 0.09410919540229885, 'recall': 0.13519091847265222, 'f1': 0.1109699279966116, 'number': 969}, 'overall_precision': 0.08680753854940035, 'overall_recall': 0.08477412158393753, 'overall_f1': 0.08577878103837472, 'overall_accuracy': 0.5670652514667025}
			------------EPOCH 4---------------
Loss:  tensor(974.9545, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1199.8013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1956.8640, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1401.8291, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(920.4337, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(976.0009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3332.7754, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1264.9141, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1715.3287, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1452.0194, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(882.8264, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(662.7694, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(732.5854, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(777.6518, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1648.3503, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1884.9249, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1004.7653, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(852.9772, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(823.1267, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1146.7515, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1664.4160, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1920.9924, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1417.2920, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1705.2043, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1103.9915, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1011.7780, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1250.2848, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(747.4099, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(508.1574, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1027.7664, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1229.5735, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(666.6433, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(811.9110, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(924.8815, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1442.6996, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.06896551724137931, 'recall': 0.02669902912621359, 'f1': 0.03849518810148731, 'number': 824}, 'P': {'precision': 0.17225806451612904, 'recall': 0.2755417956656347, 'f1': 0.21198888447796746, 'number': 969}, 'overall_precision': 0.15462814339218833, 'overall_recall': 0.1611823759063023, 'overall_f1': 0.1578372474057892, 'overall_accuracy': 0.5583546061176049}
			------------EPOCH 5---------------
Loss:  tensor(751.2375, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(952.8606, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1900.6226, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1314.9683, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(813.8867, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(878.4144, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3220.2656, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1107.2850, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1549.1354, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1300.7388, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(719.8977, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(549.7622, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(611.0223, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(687.0738, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1348.9187, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1684.3289, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(907.2013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(740.9861, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(733.9944, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1004.5863, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1486.4614, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1816.8142, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1255.4143, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1556.8724, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(962.4514, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(876.4102, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1147.7515, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(658.7532, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(393.4752, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(854.9595, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1031.3341, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(501.8505, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(643.7927, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(726.8643, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1198.8617, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.11448598130841121, 'recall': 0.05946601941747573, 'f1': 0.07827476038338659, 'number': 824}, 'P': {'precision': 0.1947019867549669, 'recall': 0.30340557275541796, 'f1': 0.2371924162968939, 'number': 969}, 'overall_precision': 0.17698658410732715, 'overall_recall': 0.1912994980479643, 'overall_f1': 0.18386491557223267, 'overall_accuracy': 0.5824712257602221}
			------------EPOCH 6---------------
Loss:  tensor(584.2046, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(789.4860, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1494.4128, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1090.2778, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(722.6786, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(800.3607, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2752.2710, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(886.0992, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1404.9844, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1290.8654, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(608.7170, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(460.3187, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(438.9398, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(672.9380, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1152.3433, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1622.3760, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(664.2123, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(582.4158, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(615.6405, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(929.3094, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1185.2854, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1488.2834, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(914.1063, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1196.8826, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(801.3591, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(748.3719, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1004.9904, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(568.7559, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(334.8941, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(836.8946, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(963.6874, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(406.0683, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(545.3967, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(602.3151, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1070.8024, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.16079494128274616, 'recall': 0.21601941747572814, 'f1': 0.18436043500776797, 'number': 824}, 'P': {'precision': 0.1303680981595092, 'recall': 0.08771929824561403, 'f1': 0.10487353485502776, 'number': 969}, 'overall_precision': 0.14951677089255258, 'overall_recall': 0.14668153931957614, 'overall_f1': 0.1480855855855856, 'overall_accuracy': 0.5777688208159792}
			------------EPOCH 7---------------
Loss:  tensor(600.4370, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(792.0663, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1325.5132, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1054.4207, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(644.6725, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(639.5023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2287.1650, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(825.3330, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1069.4100, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(938.3450, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(452.3990, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(364.7623, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(320.1476, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(581.4601, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(947.6595, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1551.2549, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(585.0724, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(601.6060, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(657.7258, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1138.4092, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1278.9064, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1640.1288, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(896.7798, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1336.0931, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(689.8480, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(630.1042, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(719.6149, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(442.6905, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(291.8210, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(574.2833, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(764.7747, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(332.6664, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(505.5521, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(447.8740, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(859.8702, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.16270783847980996, 'recall': 0.1662621359223301, 'f1': 0.16446578631452582, 'number': 824}, 'P': {'precision': 0.08412483039348712, 'recall': 0.06398348813209494, 'f1': 0.07268464243845252, 'number': 969}, 'overall_precision': 0.1260291323622546, 'overall_recall': 0.11098717233686559, 'overall_f1': 0.11803084223013048, 'overall_accuracy': 0.5600116440503381}
			------------EPOCH 8---------------
Loss:  tensor(499.8752, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(658.4047, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1160.3271, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(904.7165, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(630.2870, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(615.4918, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2056.3643, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(726.7754, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(986.9817, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(946.2406, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(503.4509, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(405.7195, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(328.3416, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(618.3219, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1265.1013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1371.2065, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(472.3711, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(410.5909, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(518.9279, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(731.0795, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(849.7538, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1163.2262, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(648.1259, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(844.1110, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(541.5784, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(511.4607, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(566.9976, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(343.2886, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(288.1542, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(612.3384, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(814.7987, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(331.6828, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(566.3370, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(534.5300, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(958.2311, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.27177700348432055, 'recall': 0.09466019417475728, 'f1': 0.1404140414041404, 'number': 824}, 'P': {'precision': 0.16791044776119404, 'recall': 0.2786377708978328, 'f1': 0.20954598370197908, 'number': 969}, 'overall_precision': 0.18364116094986807, 'overall_recall': 0.19408812046848856, 'overall_f1': 0.18872017353579176, 'overall_accuracy': 0.5905548837834207}
			------------EPOCH 9---------------
Loss:  tensor(357.1299, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(523.4092, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1203.1069, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(885.5076, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(572.6439, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(566.9342, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2601.4570, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(770.1219, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(988.7998, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(834.3715, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(335.3325, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(295.9936, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(210.1628, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(427.4664, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(858.2904, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1185.7485, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(343.6553, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(323.3915, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(385.5642, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(584.8240, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(817.8013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1156.3467, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(735.2693, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(910.0970, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(534.5497, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(502.3559, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(752.4554, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(396.2235, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(261.9072, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(580.9985, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(662.6699, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(233.9029, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(286.5328, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(332.9381, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(646.7219, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.1935483870967742, 'recall': 0.22572815533980584, 'f1': 0.20840336134453782, 'number': 824}, 'P': {'precision': 0.14258734655335223, 'recall': 0.15583075335397317, 'f1': 0.14891518737672585, 'number': 969}, 'overall_precision': 0.16683168316831684, 'overall_recall': 0.1879531511433352, 'overall_f1': 0.1767637031209022, 'overall_accuracy': 0.6102601997402481}
			------------EPOCH 10---------------
Loss:  tensor(288.6027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(477.4993, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(790.9911, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(596.3136, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(419.2178, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(387.6281, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1341.4927, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(444.3864, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(664.7892, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(580.0889, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(239.9146, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(227.7949, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(171.7994, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(326.6028, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(656.2281, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1007.4564, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(253.8575, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(325.0996, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(406.9442, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(658.7550, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(652.8690, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(929.5681, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(491.5855, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(669.2786, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(343.1323, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(334.8238, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(350.2215, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(202.3307, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(206.4434, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(341.6018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(479.3754, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(175.1809, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(270.2523, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(257.9991, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(526.3262, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.23439490445859873, 'recall': 0.22330097087378642, 'f1': 0.22871348663766317, 'number': 824}, 'P': {'precision': 0.2223926380368098, 'recall': 0.29927760577915374, 'f1': 0.2551693796744391, 'number': 969}, 'overall_precision': 0.22690282431785544, 'overall_recall': 0.2643614054656999, 'overall_f1': 0.24420401854714066, 'overall_accuracy': 0.6180303641004972}
			------------EPOCH 11---------------
Loss:  tensor(266.2751, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(458.9946, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(929.1730, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(632.6461, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(365.7753, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(421.6923, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1406.8633, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(374.4318, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(479.9263, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(420.0340, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(245.5546, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(177.9276, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(118.4027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(285.0792, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(541.5989, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(949.0872, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(177.6541, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(219.5377, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(221.6474, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(432.0533, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(526.4044, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(864.6182, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(450.7986, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(585.3164, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(409.3885, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(343.1165, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(400.1881, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(253.5620, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(171.1469, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(345.7643, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(508.0706, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(163.2511, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(212.8457, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(215.3832, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(439.0702, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.2352185089974293, 'recall': 0.2220873786407767, 'f1': 0.22846441947565543, 'number': 824}, 'P': {'precision': 0.1757142857142857, 'recall': 0.25386996904024767, 'f1': 0.2076825664837484, 'number': 969}, 'overall_precision': 0.19696969696969696, 'overall_recall': 0.2392638036809816, 'overall_f1': 0.21606648199445982, 'overall_accuracy': 0.6384298445967128}
			------------EPOCH 12---------------
Loss:  tensor(141.0270, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(287.6957, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(544.1652, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(395.2820, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(279.1108, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(257.5887, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1093.4540, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(296.2452, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(403.1936, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(361.0179, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(204.9937, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(143.8019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(111.9068, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(276.6305, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(500.1433, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(928.0079, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(139.9866, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(187.1729, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(271.0134, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(511.5348, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(532.0366, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(782.4134, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(455.2651, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(572.0747, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(244.9305, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(213.5394, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(280.1155, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(134.4630, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(146.0817, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(239.6192, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(382.0436, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(128.9654, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(160.8975, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(162.1457, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(352.6354, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.1936, 'recall': 0.14684466019417475, 'f1': 0.16701173222912352, 'number': 824}, 'P': {'precision': 0.12818532818532818, 'recall': 0.17131062951496387, 'f1': 0.14664310954063606, 'number': 969}, 'overall_precision': 0.14947916666666666, 'overall_recall': 0.16006692693809257, 'overall_f1': 0.1545919741448963, 'overall_accuracy': 0.6301222625285503}
			------------EPOCH 13---------------
Loss:  tensor(148.8145, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(250.1674, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(486.0424, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(370.1528, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(303.7539, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(260.9883, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1101.0302, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(272.3458, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(361.9313, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(293.1583, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(137.1921, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(136.2623, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(83.6710, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(206.5665, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(391.3633, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(776.4672, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(94.3929, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(143.4885, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(151.2873, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(329.1226, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(331.5342, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(637.9281, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(281.5911, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(342.3600, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(216.9511, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(177.5027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(241.7278, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(102.0811, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(131.6425, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(224.9712, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(349.2218, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(109.5775, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(103.3827, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(154.7779, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(357.0522, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.22261072261072262, 'recall': 0.23179611650485438, 'f1': 0.22711058263971465, 'number': 824}, 'P': {'precision': 0.1842680262199563, 'recall': 0.26109391124871, 'f1': 0.21605465414175917, 'number': 969}, 'overall_precision': 0.1990138951142985, 'overall_recall': 0.2476296709425544, 'overall_f1': 0.220675944333996, 'overall_accuracy': 0.6292265663486946}
			------------EPOCH 14---------------
Loss:  tensor(97.4802, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(223.4275, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(398.9610, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(336.7739, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(195.9032, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(169.5056, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(881.3615, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(222.7799, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(285.9398, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(199.8320, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(102.1523, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(92.2659, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.2212, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(171.5414, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(310.4290, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(615.7789, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(75.9275, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(121.6877, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(117.8252, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(241.2450, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(272.6439, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(571.6802, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(214.4146, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(255.2912, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(171.8652, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(156.9416, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(141.8956, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(73.1423, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(98.7434, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(158.3820, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(255.9245, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(83.0244, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(72.1787, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(94.3721, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(178.3692, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.21292775665399238, 'recall': 0.20388349514563106, 'f1': 0.20830750154990701, 'number': 824}, 'P': {'precision': 0.13793103448275862, 'recall': 0.18575851393188855, 'f1': 0.158311345646438, 'number': 969}, 'overall_precision': 0.166189111747851, 'overall_recall': 0.19408812046848856, 'overall_f1': 0.17905839979418572, 'overall_accuracy': 0.6574186036096556}
			------------EPOCH 15---------------
Loss:  tensor(65.7990, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(144.3575, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(271.5570, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(245.5057, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(146.1805, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(129.0770, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(689.0347, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(162.3065, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(222.7186, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(154.6792, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(78.2640, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(69.5918, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.9702, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(146.4442, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(263.2256, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(511.7004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.6145, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(101.1448, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(96.6828, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(188.8394, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(211.0421, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(521.3513, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(162.7231, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(209.5230, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(125.8135, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(150.4429, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(109.5948, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.4882, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(88.8419, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(117.7682, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(212.5453, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.6711, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.4431, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(65.5023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(139.4468, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.22703818369453044, 'recall': 0.2669902912621359, 'f1': 0.24539877300613497, 'number': 824}, 'P': {'precision': 0.17120622568093385, 'recall': 0.22703818369453044, 'f1': 0.19520851818988463, 'number': 969}, 'overall_precision': 0.19520851818988466, 'overall_recall': 0.24539877300613497, 'overall_f1': 0.21744502100321228, 'overall_accuracy': 0.6390344395181154}
			------------EPOCH 16---------------
Loss:  tensor(43.5011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(112.0613, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(244.2802, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(216.0829, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(104.2979, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(101.6214, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(587.2018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(117.5364, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(162.8168, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(104.7854, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(67.1613, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.6894, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.2529, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(131.4482, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(243.5717, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(446.1833, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.5690, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(77.7296, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(79.9014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(124.7907, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(184.1563, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(455.1331, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(122.2519, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(163.6051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(96.0697, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(70.1369, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(144.8401, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.3352, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.6136, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(84.9236, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(160.0910, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.4596, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.7057, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.2240, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(100.4966, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.22585438335809807, 'recall': 0.18446601941747573, 'f1': 0.20307281229124916, 'number': 824}, 'P': {'precision': 0.17103448275862068, 'recall': 0.25593395252837975, 'f1': 0.20504340636626703, 'number': 969}, 'overall_precision': 0.18841262364578426, 'overall_recall': 0.22308979364194087, 'overall_f1': 0.20429009193054135, 'overall_accuracy': 0.63838505978772}
			------------EPOCH 17---------------
Loss:  tensor(34.1323, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(67.8766, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(223.0409, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(175.2515, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(80.4214, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(84.2354, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(675.5638, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(122.7821, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(152.0023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(92.0940, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.3809, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.4927, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.5998, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(156.7028, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(290.8448, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(402.3527, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.0478, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.9708, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.1418, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(95.3003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(134.0464, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(460.5698, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(92.5940, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(131.7027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(70.0264, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(64.4050, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(131.8562, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.1055, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.5874, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(71.5672, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(184.0102, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.8774, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.6915, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.7365, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(79.0869, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.18718801996672213, 'recall': 0.27305825242718446, 'f1': 0.22211253701875616, 'number': 824}, 'P': {'precision': 0.18139097744360902, 'recall': 0.19917440660474717, 'f1': 0.1898671913428431, 'number': 969}, 'overall_precision': 0.18446601941747573, 'overall_recall': 0.2331288343558282, 'overall_f1': 0.2059620596205962, 'overall_accuracy': 0.6229119082807112}
			------------EPOCH 18---------------
Loss:  tensor(24.5483, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.9939, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(374.7401, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(165.0766, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.7857, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.1727, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(826.9563, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(160.5423, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(164.1661, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(59.0597, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.7696, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.1577, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.8511, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(114.1701, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(219.1934, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(389.8777, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.4513, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.9536, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.4509, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(84.8327, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(196.2288, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(473.7121, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(82.8379, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(126.5773, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(71.7681, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(185.0943, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.1222, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.5016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.8817, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(71.0698, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(189.2768, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.8083, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.4806, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.5319, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(62.4944, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.24497991967871485, 'recall': 0.2220873786407767, 'f1': 0.23297262889879058, 'number': 824}, 'P': {'precision': 0.18326417704011064, 'recall': 0.27347781217750256, 'f1': 0.21946169772256727, 'number': 969}, 'overall_precision': 0.20428636570907432, 'overall_recall': 0.2498605688789738, 'overall_f1': 0.22478675363773207, 'overall_accuracy': 0.634869452281786}
			------------EPOCH 19---------------
Loss:  tensor(19.6750, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.5095, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(159.9504, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(145.6409, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.4044, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.9342, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(535.5229, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(95.8864, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(125.6522, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.9877, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.4866, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.6319, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.4567, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(110.9023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(215.7450, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(369.0034, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.5443, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.9496, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.3538, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.0112, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(133.8122, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(406.7367, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.9699, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(113.3893, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.8495, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.3926, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.0651, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.7024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.1376, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.3970, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(103.8858, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.4463, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.0732, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.2316, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.9532, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.19821605550049554, 'recall': 0.24271844660194175, 'f1': 0.2182214948172395, 'number': 824}, 'P': {'precision': 0.14870180959874116, 'recall': 0.19504643962848298, 'f1': 0.16875, 'number': 969}, 'overall_precision': 0.1706140350877193, 'overall_recall': 0.2169548243167875, 'overall_f1': 0.191013994598576, 'overall_accuracy': 0.6436248824398764}
			------------EPOCH 20---------------
Loss:  tensor(15.8397, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.5236, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(134.3812, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(112.6384, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.0632, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.9636, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(436.7524, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(82.0233, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(93.8092, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.1421, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.0862, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.8842, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.5166, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(100.2354, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(193.8844, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(326.4836, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.6534, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.4640, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.0313, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.3194, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(77.7899, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(366.7249, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.0142, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(80.7665, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.3504, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.2233, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.2535, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.7163, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.6231, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.0785, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(88.4288, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.0082, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.2496, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.7371, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.6288, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.24114441416893734, 'recall': 0.21480582524271843, 'f1': 0.22721437740693193, 'number': 824}, 'P': {'precision': 0.18565683646112602, 'recall': 0.28586171310629516, 'f1': 0.22511174319382365, 'number': 969}, 'overall_precision': 0.20395327942497754, 'overall_recall': 0.2532069157836029, 'overall_f1': 0.22592684747449612, 'overall_accuracy': 0.6401092749339424}
			------------EPOCH 21---------------
Loss:  tensor(10.0923, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.8096, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(120.3582, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(93.4512, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.8479, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.5735, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(371.9435, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(67.0206, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(70.3661, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.6131, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.1375, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.7188, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.0444, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(80.0914, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(155.5352, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(289.6815, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.1560, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.6844, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.8952, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.0419, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(66.7945, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(309.1783, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.3414, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.3246, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.5114, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.7951, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.6025, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.3973, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.7737, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.7542, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(66.4309, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.2095, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.3807, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.5769, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.3122, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.21428571428571427, 'recall': 0.24393203883495146, 'f1': 0.22814982973893305, 'number': 824}, 'P': {'precision': 0.18521400778210118, 'recall': 0.24561403508771928, 'f1': 0.2111801242236025, 'number': 969}, 'overall_precision': 0.197480881691408, 'overall_recall': 0.24484104852203012, 'overall_f1': 0.21862549800796813, 'overall_accuracy': 0.6373550091808858}
			------------EPOCH 22---------------
Loss:  tensor(8.0181, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.0501, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(97.3649, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(77.2577, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.4456, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.4916, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(330.7867, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.9120, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.9307, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.3483, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.9053, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.6263, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.0552, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(66.8329, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(128.6031, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(259.4443, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.7057, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.4891, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.5062, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.2269, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.5840, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(274.3889, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.8173, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.8454, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.0165, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.5549, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.5926, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.7116, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.0719, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.2536, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.4565, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.8683, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.1079, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.3308, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.8880, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.20964125560538116, 'recall': 0.22694174757281554, 'f1': 0.21794871794871795, 'number': 824}, 'P': {'precision': 0.15927272727272726, 'recall': 0.2260061919504644, 'f1': 0.18686006825938564, 'number': 969}, 'overall_precision': 0.17909131010145568, 'overall_recall': 0.22643614054657, 'overall_f1': 0.2, 'overall_accuracy': 0.6376685028438354}
			------------EPOCH 23---------------
Loss:  tensor(6.2288, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.9389, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(91.5957, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.7602, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.7900, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.3689, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(286.1273, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.3853, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.6560, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.5948, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.4801, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.0847, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.7003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.2307, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(100.5697, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(238.4566, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.7572, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.7104, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.2986, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.4501, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.5383, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(244.5741, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.3111, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.1461, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.6159, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.4709, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.1551, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.4368, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.0198, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.6585, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.7423, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.7390, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.1714, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.5372, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.9525, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.2043956043956044, 'recall': 0.22572815533980584, 'f1': 0.21453287197231835, 'number': 824}, 'P': {'precision': 0.1816126601356443, 'recall': 0.24871001031991744, 'f1': 0.20993031358885017, 'number': 969}, 'overall_precision': 0.19088064371926688, 'overall_recall': 0.2381483547127719, 'overall_f1': 0.21191066997518612, 'overall_accuracy': 0.6384746294057055}
			------------EPOCH 24---------------
Loss:  tensor(4.9384, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.8617, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(77.4002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.4599, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.0723, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.0061, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(255.3718, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.2195, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.9397, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.9996, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.6289, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.3558, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.3620, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.6612, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(79.3500, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(214.8347, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.6515, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.8861, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.2548, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.2552, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(72.1741, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(220.9717, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.6973, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.9787, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.8300, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.9321, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.1112, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.5637, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.5838, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.4492, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.1333, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.3528, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.8444, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.1855, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.5917, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.20521541950113378, 'recall': 0.2196601941747573, 'f1': 0.21219226260257912, 'number': 824}, 'P': {'precision': 0.1786492374727669, 'recall': 0.25386996904024767, 'f1': 0.20971867007672637, 'number': 969}, 'overall_precision': 0.18902169101372288, 'overall_recall': 0.2381483547127719, 'overall_f1': 0.21076011846001977, 'overall_accuracy': 0.6354964396076851}
			------------EPOCH 25---------------
Loss:  tensor(4.0945, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.6070, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(73.6756, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.7903, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.2183, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.2813, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(252.0178, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.1472, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.0085, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.7725, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.1112, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.4152, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.7706, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.3515, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(87.2495, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(206.5957, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.0463, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.6664, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.5578, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.1490, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(224.0259, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(243.6992, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.4028, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.1055, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.5442, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.7752, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.4866, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.7240, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.9557, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.2806, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.0149, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.1763, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.2044, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.1859, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.5030, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.2025028441410694, 'recall': 0.21601941747572814, 'f1': 0.20904286553141516, 'number': 824}, 'P': {'precision': 0.14654517843583903, 'recall': 0.19917440660474717, 'f1': 0.16885389326334208, 'number': 969}, 'overall_precision': 0.16894353369763207, 'overall_recall': 0.20691578360290017, 'overall_f1': 0.1860115317122086, 'overall_accuracy': 0.6459089076985086}
			------------EPOCH 26---------------
Loss:  tensor(5.1144, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.6096, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(80.9654, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(124.6710, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.1776, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.8037, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(312.5498, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.7778, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.0835, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.0427, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.2835, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.7967, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.6813, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(89.1982, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(179.2756, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(228.2562, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.9542, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.9643, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.0988, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.3587, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(129.0785, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(328.4947, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(62.8403, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.2000, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.3134, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(72.1061, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.9640, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.8337, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.6326, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.1930, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.8117, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.8256, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.0102, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.8801, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.5022, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.2388955582232893, 'recall': 0.24150485436893204, 'f1': 0.24019312009656005, 'number': 824}, 'P': {'precision': 0.18449931412894377, 'recall': 0.2776057791537668, 'f1': 0.22167284713638236, 'number': 969}, 'overall_precision': 0.20427760803142733, 'overall_recall': 0.2610150585610708, 'overall_f1': 0.22918707149853085, 'overall_accuracy': 0.6339737561019302}
			------------EPOCH 27---------------
Loss:  tensor(4.1076, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.9983, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(74.3801, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(144.2738, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.6887, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.9650, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(368.8995, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.6245, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.0495, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.3381, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.7993, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.6628, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.0086, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(92.2739, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(184.1992, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(252.1513, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.6679, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.1112, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.8568, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.6852, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(67.5373, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(255.3484, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.3004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.9441, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.6119, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.2503, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(62.3311, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.5160, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.1724, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.1754, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.1647, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.0431, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.3820, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.0505, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.2025, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.22426095820591233, 'recall': 0.2669902912621359, 'f1': 0.24376731301939059, 'number': 824}, 'P': {'precision': 0.176, 'recall': 0.22703818369453044, 'f1': 0.19828751689950427, 'number': 969}, 'overall_precision': 0.19722097714029582, 'overall_recall': 0.24539877300613497, 'overall_f1': 0.21868787276341947, 'overall_accuracy': 0.6392583635630794}
			------------EPOCH 28---------------
Loss:  tensor(4.9314, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.1791, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(73.6169, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(71.7667, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.1978, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.4412, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(271.0420, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.8944, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.7141, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.4822, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.7030, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.3007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.1551, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.9166, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(103.0726, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(253.5371, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.5569, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.4755, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.0139, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.2959, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.4211, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(218.9909, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.5368, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.7759, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.6274, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.1767, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.2806, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.9054, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.0080, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.6854, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.9751, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.8791, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.8543, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.4570, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.1847, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.23037323037323038, 'recall': 0.21723300970873785, 'f1': 0.22361024359775136, 'number': 824}, 'P': {'precision': 0.1631504922644163, 'recall': 0.239422084623323, 'f1': 0.19406106231702216, 'number': 969}, 'overall_precision': 0.1869031377899045, 'overall_recall': 0.22922476296709426, 'overall_f1': 0.20591182364729457, 'overall_accuracy': 0.6406690850463522}
			------------EPOCH 29---------------
Loss:  tensor(4.2460, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.7490, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(67.2457, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.2475, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.9050, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.2180, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(219.8393, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.0022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.5769, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.6030, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.4795, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.5923, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.4536, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.0440, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(106.0151, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(231.1416, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.7018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.0778, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.8609, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.2091, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.1899, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(190.9893, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.2169, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.9846, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.2965, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.3675, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.4039, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.4932, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.3296, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.2881, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.1431, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.0086, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.5922, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.3407, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.7609, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.21790722761596548, 'recall': 0.24514563106796117, 'f1': 0.23072529982866932, 'number': 824}, 'P': {'precision': 0.17686567164179104, 'recall': 0.24458204334365324, 'f1': 0.20528367258553484, 'number': 969}, 'overall_precision': 0.19364799294221438, 'overall_recall': 0.24484104852203012, 'overall_f1': 0.21625615763546796, 'overall_accuracy': 0.6415423888217117}
			------------EPOCH 30---------------
Loss:  tensor(2.6898, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.5926, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.9901, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.6972, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.4615, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.8894, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(206.5151, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.1083, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.0032, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.7337, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.7892, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.5088, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.7765, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.4938, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(65.7533, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(184.8582, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.8645, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.8207, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.0944, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.0437, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.6481, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(155.3727, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.6631, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.5305, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.5013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.9209, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.0864, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.0351, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.5145, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.7614, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.2901, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.3358, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.9591, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.3560, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.8119, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.22168087697929353, 'recall': 0.220873786407767, 'f1': 0.2212765957446809, 'number': 824}, 'P': {'precision': 0.18347338935574228, 'recall': 0.27038183694530443, 'f1': 0.21860659157279935, 'number': 969}, 'overall_precision': 0.1974210760337928, 'overall_recall': 0.2476296709425544, 'overall_f1': 0.21969322117763487, 'overall_accuracy': 0.6400197053159569}


		-------------RUN 5-----------
			------------EPOCH 1---------------
Loss:  tensor(5064.0625, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1999.9250, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3436.2913, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1705.8633, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1350.8759, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1641.2903, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(743.8551, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2758.7561, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2061.1423, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2196.3806, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3045.2847, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1302.9443, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1576.1725, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1297.7169, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2068.3855, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1074.5854, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1734.4380, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1382.6118, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2612.1306, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1850.5758, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1266.7859, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1347.3566, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2263.4109, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2246.5073, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3052.2856, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2362.4585, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1543.0945, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1470.4139, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1115.9058, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1864.2366, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2129.5649, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1699.0913, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2210.1079, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1287.9657, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1463.6934, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2181.1226, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2169.2256, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2516.3811, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.009259259259259259, 'recall': 0.004392386530014641, 'f1': 0.005958291956305858, 'number': 683}, 'P': {'precision': 0.051351351351351354, 'recall': 0.02261904761904762, 'f1': 0.031404958677685946, 'number': 840}, 'overall_precision': 0.03170028818443804, 'overall_recall': 0.014445173998686802, 'overall_f1': 0.019846639603067207, 'overall_accuracy': 0.4506957836034815}
			------------EPOCH 2---------------
Loss:  tensor(3801.1477, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1517.8582, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2475.0085, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1286.0803, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1119.0295, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1345.9146, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(641.6371, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2233.8652, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1508.7148, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1737.8765, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2498.2141, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1066.1969, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1355.0881, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1135.5911, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1843.5901, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(909.4390, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1487.7443, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1197.6526, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2339.5764, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1568.4481, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1087.5453, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1165.3345, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2075.3772, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1999.6433, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2677.7510, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2020.6271, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1328.9249, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1246.5581, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(911.4260, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1588.5315, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1674.5190, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1433.3364, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1855.4622, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1006.3660, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1228.2588, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1721.1689, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1853.2140, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2181.6860, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.030959752321981424, 'recall': 0.014641288433382138, 'f1': 0.019880715705765408, 'number': 683}, 'P': {'precision': 0.06385869565217392, 'recall': 0.055952380952380955, 'f1': 0.05964467005076143, 'number': 840}, 'overall_precision': 0.053824362606232294, 'overall_recall': 0.03742613263296126, 'overall_f1': 0.04415182029434547, 'overall_accuracy': 0.5115182154583833}
			------------EPOCH 3---------------
Loss:  tensor(3571.5459, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1381.0774, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2294.4651, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1128.5542, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(998.0104, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1202.8646, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(552.2351, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1951.4397, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1267.0500, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1429.4747, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2153.3979, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(922.6193, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1109.7751, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1003.2932, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1589.9709, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(755.6390, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1282.2034, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(994.5955, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2007.1787, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1330.3135, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(957.2745, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(994.9091, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1712.5171, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1735.5602, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2272.3984, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1754.7158, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1016.3015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(916.6360, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(681.1672, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1242.9066, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1350.0930, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1213.7310, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1608.7260, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(839.6046, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1066.5254, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1408.7701, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1530.2571, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1931.9330, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.06292517006802721, 'recall': 0.05417276720351391, 'f1': 0.05822187254130606, 'number': 683}, 'P': {'precision': 0.10159651669085631, 'recall': 0.08333333333333333, 'f1': 0.09156311314584695, 'number': 840}, 'overall_precision': 0.08379013312451057, 'overall_recall': 0.07025607353906763, 'overall_f1': 0.07642857142857142, 'overall_accuracy': 0.5736957314848595}
			------------EPOCH 4---------------
Loss:  tensor(2880.4248, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1111.6094, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1864.1198, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(931.8073, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(835.4321, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(947.4702, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(428.6811, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1646.5966, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(933.2133, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1052.9384, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1801.9324, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(799.2198, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(903.3704, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(914.9415, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1288.4033, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(611.1615, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1086.4382, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(782.9160, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1657.9877, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1157.1877, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(872.9191, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(830.7563, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1237.3147, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1422.3333, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1823.0012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1448.1582, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(629.2034, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(562.7885, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(492.4481, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(882.6002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1043.2896, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(980.2332, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1391.8247, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(676.4779, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(844.9489, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1114.3916, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1227.3771, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1816.7498, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.1324582338902148, 'recall': 0.16251830161054173, 'f1': 0.14595660749506906, 'number': 683}, 'P': {'precision': 0.2234185733512786, 'recall': 0.1976190476190476, 'f1': 0.20972836386607707, 'number': 840}, 'overall_precision': 0.1752055660974067, 'overall_recall': 0.1818778726198293, 'overall_f1': 0.17847938144329897, 'overall_accuracy': 0.5914421222702871}
			------------EPOCH 5---------------
Loss:  tensor(2430.7812, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(838.4958, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1492.7722, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(791.9977, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(718.1567, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(722.8036, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(325.1402, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1366.4811, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(631.6841, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(820.6376, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1352.0054, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(650.7478, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(683.7108, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(781.1059, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1088.1650, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(485.7915, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(933.9774, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(643.9742, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1548.6360, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1000.0524, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(733.5817, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(682.6157, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(817.6720, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1177.7908, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1477.9268, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1249.3983, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(439.7554, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(405.9104, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(450.3535, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(682.9305, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(890.7677, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(810.4642, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1133.1958, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(536.5154, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(732.8911, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(916.0672, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(945.6754, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1542.6379, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.17370892018779344, 'recall': 0.2708638360175695, 'f1': 0.2116704805491991, 'number': 683}, 'P': {'precision': 0.2254335260115607, 'recall': 0.18571428571428572, 'f1': 0.20365535248041777, 'number': 840}, 'overall_precision': 0.19408081957882756, 'overall_recall': 0.22390019697964544, 'overall_f1': 0.2079268292682927, 'overall_accuracy': 0.5694741231041851}
			------------EPOCH 6---------------
Loss:  tensor(2075.1594, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(700.6777, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1189.6927, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(736.0768, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(644.7922, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(620.8505, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(286.4771, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1315.2539, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(559.5891, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(678.3831, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1081.4307, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(604.1872, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(561.6115, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(610.8230, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(933.2181, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(367.9096, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(722.2325, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(446.9854, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1106.1526, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(823.1250, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(605.8273, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(588.3182, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(590.6180, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1103.5598, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1367.2253, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1205.3143, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(388.2096, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(410.0101, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(369.3196, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(616.3357, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(781.2382, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(689.4108, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(848.7734, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(401.1812, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(541.7043, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(755.1016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(803.6406, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1226.4607, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.154434250764526, 'recall': 0.1478770131771596, 'f1': 0.15108451757666416, 'number': 683}, 'P': {'precision': 0.20257234726688103, 'recall': 0.3, 'f1': 0.2418426103646833, 'number': 840}, 'overall_precision': 0.18598524762908325, 'overall_recall': 0.23177938279711097, 'overall_f1': 0.2063724057293189, 'overall_accuracy': 0.6082503778600094}
			------------EPOCH 7---------------
Loss:  tensor(1811.0935, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(507.5284, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1058.9929, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(696.1096, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(497.2327, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(480.0339, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(236.4173, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(988.5192, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(631.1261, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(687.5437, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1048.7827, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(614.1221, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(751.8771, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(580.1151, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1151.8689, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(472.9519, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(731.8337, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(443.1096, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1147.9634, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(668.3433, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(434.8974, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(472.6110, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(656.0363, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(807.7831, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(977.8105, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(957.3930, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(253.0243, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(274.6865, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(234.1445, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(366.7687, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(550.4055, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(439.2176, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(706.2491, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(321.7870, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(462.1190, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(614.8302, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(706.3977, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1050.0283, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.2986425339366516, 'recall': 0.09663250366032211, 'f1': 0.14601769911504425, 'number': 683}, 'P': {'precision': 0.16654492330168005, 'recall': 0.2714285714285714, 'f1': 0.20642824807605248, 'number': 840}, 'overall_precision': 0.18490566037735848, 'overall_recall': 0.19304005252790546, 'overall_f1': 0.18888531962736912, 'overall_accuracy': 0.6111429613801012}
			------------EPOCH 8---------------
Loss:  tensor(1956.9316, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(641.3854, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1330.3712, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(939.5616, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(554.0927, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(680.9362, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(423.4791, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1042.0072, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(337.7987, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(514.8016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(865.4900, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(343.0740, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(355.5074, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(476.5057, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(689.7509, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(228.8119, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(570.4532, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(295.2381, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(809.1240, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(620.8671, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(465.9706, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(492.6117, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(482.1049, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1104.8853, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1779.0983, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1874.4790, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(270.5016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(257.0548, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(132.1433, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(323.4020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(715.2373, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(452.9634, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(705.4955, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(246.8774, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(481.0907, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(530.8499, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(788.4811, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1414.6738, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.22590361445783133, 'recall': 0.10980966325036604, 'f1': 0.1477832512315271, 'number': 683}, 'P': {'precision': 0.20643086816720257, 'recall': 0.3821428571428571, 'f1': 0.2680584551148225, 'number': 840}, 'overall_precision': 0.20985691573926868, 'overall_recall': 0.26001313197636244, 'overall_f1': 0.23225806451612904, 'overall_accuracy': 0.5735654349299004}
			------------EPOCH 9---------------
Loss:  tensor(2070.5239, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(462.9411, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1082.2289, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(756.5093, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(513.0713, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(611.3150, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(318.8624, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1280.4845, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(356.8626, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(501.4818, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(885.2565, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(460.7822, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(425.0261, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(461.3024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(627.2822, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(227.8547, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(493.9785, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(294.1990, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(732.5699, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(543.1287, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(387.6221, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(497.1119, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(340.5673, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(762.8907, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1478.6267, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1513.2852, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(342.0363, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(327.9846, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(276.0818, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(533.9056, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1093.7954, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(925.3872, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1225.9792, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(650.5436, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1022.2520, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1240.8394, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(994.8688, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1356.5977, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.2105843439911797, 'recall': 0.2796486090775988, 'f1': 0.24025157232704403, 'number': 683}, 'P': {'precision': 0.19615912208504802, 'recall': 0.17023809523809524, 'f1': 0.18228170809432762, 'number': 840}, 'overall_precision': 0.2041564792176039, 'overall_recall': 0.21930400525279053, 'overall_f1': 0.2114593225704337, 'overall_accuracy': 0.5959764423828634}
			------------EPOCH 10---------------
Loss:  tensor(1670.8960, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(486.4310, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(948.5659, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(476.7317, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(333.2496, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(349.5261, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(181.2015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(655.2119, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(390.8252, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(476.2012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(733.5691, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(323.8828, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(760.7373, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(821.4352, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1274.8191, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(494.1197, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1078.0205, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(857.5217, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1755.4478, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(977.4587, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(632.3539, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(693.5135, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1379.8843, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1167.1301, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1110.0367, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(910.3672, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(209.9959, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(182.8955, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(140.7766, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(294.5618, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(573.3403, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(320.2385, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(703.3679, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(230.2215, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(465.6917, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(485.2755, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(658.8420, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1080.2800, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.1751412429378531, 'recall': 0.31771595900439237, 'f1': 0.2258064516129032, 'number': 683}, 'P': {'precision': 0.09815950920245399, 'recall': 0.05714285714285714, 'f1': 0.07223476297968397, 'number': 840}, 'overall_precision': 0.15335648148148148, 'overall_recall': 0.17399868680236374, 'overall_f1': 0.16302676099661642, 'overall_accuracy': 0.5457080314796476}
			------------EPOCH 11---------------
Loss:  tensor(1575.6199, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(485.5043, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(976.6482, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(559.9550, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(494.9787, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(496.7236, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(222.1738, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(952.9895, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(530.4906, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(661.8483, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1086.5833, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(412.3961, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(366.5430, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(430.9510, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(654.2106, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(273.3557, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(433.5722, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(250.3655, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(676.9440, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(565.6381, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(279.9358, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(335.4299, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(294.2598, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(580.8109, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(722.8225, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(777.2747, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(166.1954, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(179.3819, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(140.7291, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(241.8729, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(499.1523, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(306.6641, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(486.0056, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(175.3233, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(373.9194, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(363.4486, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(483.2633, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(744.8127, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.229706390328152, 'recall': 0.19472913616398244, 'f1': 0.21077654516640254, 'number': 683}, 'P': {'precision': 0.15924092409240925, 'recall': 0.22976190476190475, 'f1': 0.18810916179337234, 'number': 840}, 'overall_precision': 0.1820212171970966, 'overall_recall': 0.21405121470781352, 'overall_f1': 0.19674109837054918, 'overall_accuracy': 0.6236774899671653}
			------------EPOCH 12---------------
Loss:  tensor(1095.4287, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(282.3492, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(650.4231, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(448.6425, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(241.5056, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(288.1526, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(156.6693, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(476.9756, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(189.3008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(291.6269, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(382.3122, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(177.6869, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(218.7061, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(332.8572, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(497.1354, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(155.7047, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(321.2702, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(225.4699, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(490.2650, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(333.1081, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(256.0877, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(278.5081, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(347.5742, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(365.9778, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(522.3306, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(633.7174, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(130.1061, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(125.9291, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(80.8493, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(158.3926, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(404.2672, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(201.3889, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(363.2507, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(132.0890, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(323.8069, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(293.6744, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(352.2032, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(617.3554, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.2089227421109902, 'recall': 0.28111273792093705, 'f1': 0.2397003745318352, 'number': 683}, 'P': {'precision': 0.17545748116254037, 'recall': 0.19404761904761905, 'f1': 0.1842849067269644, 'number': 840}, 'overall_precision': 0.1920995670995671, 'overall_recall': 0.2330925804333552, 'overall_f1': 0.2106199940670424, 'overall_accuracy': 0.6297493094282587}
			------------EPOCH 13---------------
Loss:  tensor(978.7685, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(242.5193, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(573.4097, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(323.4351, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(175.7964, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(214.3872, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(122.3904, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(289.6795, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(162.9444, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(285.8825, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(323.7111, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(139.3564, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(150.7820, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(253.3927, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(342.8973, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(97.5023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(255.1093, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(140.0139, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(316.6926, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(273.9521, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(155.2275, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(190.2209, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(169.1345, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(187.7478, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(421.0292, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(532.1600, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(99.4968, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(99.4734, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(62.1601, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(120.2789, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(300.1093, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(134.2476, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(286.7818, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(96.7317, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(240.9735, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(208.7486, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(287.2252, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(451.6780, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.21940298507462686, 'recall': 0.21522693997071743, 'f1': 0.2172949002217295, 'number': 683}, 'P': {'precision': 0.1760948905109489, 'recall': 0.22976190476190475, 'f1': 0.19938016528925617, 'number': 840}, 'overall_precision': 0.19252548131370328, 'overall_recall': 0.2232435981615233, 'overall_f1': 0.2067497719671633, 'overall_accuracy': 0.636420493042164}
			------------EPOCH 14---------------
Loss:  tensor(811.5479, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(178.1792, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(456.0956, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(259.7336, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(131.0519, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(148.3860, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(97.9311, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(212.2957, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(110.5901, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(226.0087, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(236.0855, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(105.2821, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(123.6212, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(211.6113, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(265.4387, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(80.8384, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(187.6826, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(102.1722, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(227.4290, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(204.5797, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(121.8652, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(157.2404, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(170.5754, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(125.9959, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(362.7472, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(482.5500, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(79.1818, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(68.9856, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.8345, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(88.5766, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(230.3826, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(95.5450, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(221.6030, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(73.5258, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(209.2990, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(169.5178, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(228.9351, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(391.2825, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.20287958115183247, 'recall': 0.22693997071742314, 'f1': 0.21423635107118177, 'number': 683}, 'P': {'precision': 0.16801437556154536, 'recall': 0.2226190476190476, 'f1': 0.19150025601638504, 'number': 840}, 'overall_precision': 0.1822056473095365, 'overall_recall': 0.22455679579776758, 'overall_f1': 0.20117647058823532, 'overall_accuracy': 0.6355865950904258}
			------------EPOCH 15---------------
Loss:  tensor(647.8776, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(141.8116, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(373.0038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(194.3948, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(102.7002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(112.5185, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(75.9715, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(162.6778, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(91.1885, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(162.4124, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(194.5631, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(87.9459, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(85.7149, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(157.8078, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(201.0345, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.3755, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(141.7754, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(71.1126, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(175.3201, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(164.0404, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(89.0442, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(131.9169, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(98.0712, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(98.7505, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(314.0251, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(433.3400, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(68.2126, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.9003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.2537, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(65.6425, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(201.6980, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(70.7888, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(178.6384, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(61.5028, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(159.6203, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(129.7635, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(180.8571, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(311.1153, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.24671916010498687, 'recall': 0.2752562225475842, 'f1': 0.26020761245674745, 'number': 683}, 'P': {'precision': 0.18518518518518517, 'recall': 0.23214285714285715, 'f1': 0.20602218700475436, 'number': 840}, 'overall_precision': 0.21101928374655649, 'overall_recall': 0.25147734734077476, 'overall_f1': 0.22947872977831038, 'overall_accuracy': 0.641554177307552}
			------------EPOCH 16---------------
Loss:  tensor(584.3195, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(120.0158, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(322.8564, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(150.6978, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(82.4903, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(78.6611, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.2877, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(120.4096, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(62.6976, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(120.9991, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(155.2629, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(69.3871, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(78.1874, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(121.0331, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(160.1555, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.4779, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(104.1845, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.3866, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(137.6528, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(124.8172, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(69.3235, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(104.3334, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.4918, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(70.8350, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(276.1615, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(388.3630, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.8981, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.0094, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.4801, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.7281, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(175.1736, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.5033, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(145.9664, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.3028, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(118.2927, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(95.5962, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(145.8763, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(238.5676, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.21721311475409835, 'recall': 0.232796486090776, 'f1': 0.22473498233215547, 'number': 683}, 'P': {'precision': 0.1586327782646801, 'recall': 0.2154761904761905, 'f1': 0.18273599192327106, 'number': 840}, 'overall_precision': 0.1815269620928991, 'overall_recall': 0.2232435981615233, 'overall_f1': 0.20023557126030628, 'overall_accuracy': 0.6405378641788712}
			------------EPOCH 17---------------
Loss:  tensor(510.0161, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(103.4331, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(279.0983, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(114.5689, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(65.5384, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(73.1307, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.4926, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(97.5806, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.0337, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(91.7205, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(123.7179, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.1860, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(64.0479, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(91.8506, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(125.1365, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.5614, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(72.2468, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.4857, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(111.3656, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(92.0401, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.6003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(88.7025, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.7139, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.5824, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(248.8039, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(347.4039, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.3792, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.4067, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.0369, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.3846, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(150.2650, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.3079, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(126.0394, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.1092, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(93.0130, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(74.7679, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(121.6558, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(196.1627, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.23346303501945526, 'recall': 0.2635431918008785, 'f1': 0.24759284731774417, 'number': 683}, 'P': {'precision': 0.17668161434977578, 'recall': 0.23452380952380952, 'f1': 0.20153452685421996, 'number': 840}, 'overall_precision': 0.19989395546129374, 'overall_recall': 0.247537754432042, 'overall_f1': 0.22117923144617188, 'overall_accuracy': 0.6406421014228384}
			------------EPOCH 18---------------
Loss:  tensor(453.4507, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(83.7208, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(238.1563, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(88.2864, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.3825, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(71.4484, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.5409, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(73.9683, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.3448, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(73.4416, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(94.1057, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.0420, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.8889, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(74.6074, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(101.7716, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.3356, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.3440, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.0188, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(97.0746, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(71.6590, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.7982, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(74.3967, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.6529, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.3218, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(253.2857, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(320.1395, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.7658, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.3431, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.0243, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.6272, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(137.2133, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.4922, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(108.2827, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.8979, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(65.7790, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.7054, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(101.9379, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(165.7610, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.23036649214659685, 'recall': 0.25768667642752563, 'f1': 0.24326192121630957, 'number': 683}, 'P': {'precision': 0.17068811438784628, 'recall': 0.22738095238095238, 'f1': 0.19499744767738642, 'number': 840}, 'overall_precision': 0.19490175252257036, 'overall_recall': 0.24097176625082076, 'overall_f1': 0.21550205519671167, 'overall_accuracy': 0.6308698598009068}
			------------EPOCH 19---------------
Loss:  tensor(412.7164, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(70.5914, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(205.0274, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(67.8036, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.5494, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.0500, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.0558, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.1291, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.8105, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.1578, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(73.0952, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.1603, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.1238, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(62.2485, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(81.1202, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.9963, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.2401, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.2427, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(97.7667, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(62.6394, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.3947, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.0639, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.7532, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.6349, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(304.4296, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(315.0187, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.4598, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.3640, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.7445, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.7243, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(140.8427, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.4577, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(90.9029, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.6639, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.4599, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.4364, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(92.2078, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(155.9560, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.21052631578947367, 'recall': 0.2108345534407028, 'f1': 0.21068032187271396, 'number': 683}, 'P': {'precision': 0.1509433962264151, 'recall': 0.21904761904761905, 'f1': 0.1787275376396309, 'number': 840}, 'overall_precision': 0.1723594324750394, 'overall_recall': 0.2153644123440578, 'overall_f1': 0.19147694103911267, 'overall_accuracy': 0.640694220044822}
			------------EPOCH 20---------------
Loss:  tensor(411.5377, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(64.3506, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(202.4545, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.5931, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.1285, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.9227, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.7431, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.7164, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.0145, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.8267, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.9336, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.9723, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.7382, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.0340, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(68.1576, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.4246, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.4337, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.3458, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(68.6225, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.2354, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.0757, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.3129, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.2989, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.4285, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(211.1608, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(259.5372, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.1266, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.4076, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.8548, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.8557, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(112.7991, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.7036, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(70.8864, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.3642, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.7267, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.3461, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(122.6202, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(152.8727, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.21460176991150443, 'recall': 0.2840409956076135, 'f1': 0.24448645242596093, 'number': 683}, 'P': {'precision': 0.1807348560079444, 'recall': 0.21666666666666667, 'f1': 0.1970763400108284, 'number': 840}, 'overall_precision': 0.1967556253270539, 'overall_recall': 0.2468811556139199, 'overall_f1': 0.21898660454280722, 'overall_accuracy': 0.6340751550529004}
			------------EPOCH 21---------------
Loss:  tensor(660.7290, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(144.1471, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(307.3002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.1074, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.9795, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.0782, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.6942, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.9465, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.8935, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.7523, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.6548, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.6409, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.9408, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.4673, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.6611, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.0669, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.6495, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.3306, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(89.1452, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.6924, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.3520, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.3614, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.2751, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.7474, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(315.4276, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(310.3782, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.2635, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.3776, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.2059, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.9343, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(154.9351, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.0111, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(98.9426, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.3342, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.2630, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.3343, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(150.3382, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(202.2752, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.24477611940298508, 'recall': 0.24011713030746706, 'f1': 0.24242424242424243, 'number': 683}, 'P': {'precision': 0.18616144975288304, 'recall': 0.26904761904761904, 'f1': 0.2200584225900682, 'number': 840}, 'overall_precision': 0.2070063694267516, 'overall_recall': 0.2560735390676297, 'overall_f1': 0.2289404167889639, 'overall_accuracy': 0.6324334184604159}
			------------EPOCH 22---------------
Loss:  tensor(649.4203, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(129.0394, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(330.6763, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.7941, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.2113, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.4359, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.7036, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.0615, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.5985, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.7145, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.2215, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.8161, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.1649, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.0365, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.4685, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.5169, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.6278, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.1336, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(68.3351, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.2228, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.6111, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.2701, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.5118, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.9792, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(249.9479, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(283.1901, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.3541, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.2827, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.9114, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.8376, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(159.8253, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.8808, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.6109, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.9556, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.7762, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.3051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(99.1714, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(148.2530, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.1959531416400426, 'recall': 0.26939970717423134, 'f1': 0.22688039457459924, 'number': 683}, 'P': {'precision': 0.17486338797814208, 'recall': 0.19047619047619047, 'f1': 0.18233618233618235, 'number': 840}, 'overall_precision': 0.1855447680690399, 'overall_recall': 0.22586999343401182, 'overall_f1': 0.20373112229789755, 'overall_accuracy': 0.6127586386615937}
			------------EPOCH 23---------------
Loss:  tensor(439.1403, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(74.9910, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(193.7925, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.9297, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.6071, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.8020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.4664, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.6167, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.2196, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.6626, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.2577, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.6141, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.9874, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.2091, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.2456, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.3158, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.4298, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.0267, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.3880, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.2299, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.3234, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.8295, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.8879, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.4857, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(198.7665, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(226.5852, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.1109, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.9026, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.9220, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.5313, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(98.1422, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.7641, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.6150, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.8576, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.4182, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.6065, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(122.5842, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(210.8010, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.23579545454545456, 'recall': 0.2430453879941435, 'f1': 0.23936553713049752, 'number': 683}, 'P': {'precision': 0.17672047578589634, 'recall': 0.24761904761904763, 'f1': 0.2062469013386217, 'number': 840}, 'overall_precision': 0.19883040935672514, 'overall_recall': 0.24556795797767564, 'overall_f1': 0.21974148061104584, 'overall_accuracy': 0.6398863814040757}
			------------EPOCH 24---------------
Loss:  tensor(350.1686, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.6784, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(142.1663, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.5207, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.6325, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.5612, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.7033, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.7383, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.0373, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.5653, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.6935, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.3860, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.3916, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.9554, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.4018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.3041, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.3120, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.9726, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.7745, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.4245, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.5531, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.2833, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.1199, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.8019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(175.6435, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(204.5324, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.1478, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.1683, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.5078, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.1786, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(85.7009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.6521, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.7573, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.8548, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.1155, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.6678, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(80.1144, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(126.6687, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.2291139240506329, 'recall': 0.26500732064421667, 'f1': 0.2457569585879158, 'number': 683}, 'P': {'precision': 0.1704152249134948, 'recall': 0.23452380952380952, 'f1': 0.19739478957915832, 'number': 840}, 'overall_precision': 0.19424460431654678, 'overall_recall': 0.24819435325016415, 'overall_f1': 0.21793023926203517, 'overall_accuracy': 0.6323552405274404}
			------------EPOCH 25---------------
Loss:  tensor(316.2689, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.5761, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(125.9674, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.7827, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.2413, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.8655, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.1962, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.1522, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.4497, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.2830, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.2574, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.9018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.6451, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.1861, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.4160, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.2471, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.2082, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.6158, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.7547, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.4980, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.5591, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.0558, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.3053, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.5736, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(154.8336, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(177.7038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.1150, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.0652, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.1093, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.0121, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(70.9302, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.9140, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.9162, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.5359, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.2422, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.9651, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(61.8453, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(92.7191, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.20266666666666666, 'recall': 0.2225475841874085, 'f1': 0.212142358688067, 'number': 683}, 'P': {'precision': 0.16379310344827586, 'recall': 0.2261904761904762, 'f1': 0.18999999999999997, 'number': 840}, 'overall_precision': 0.17905759162303664, 'overall_recall': 0.22455679579776758, 'overall_f1': 0.19924264491698224, 'overall_accuracy': 0.637567102725804}
			------------EPOCH 26---------------
Loss:  tensor(276.3695, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.6577, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(106.4457, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.0738, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.6743, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.2811, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.7372, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.7518, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.4361, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.4982, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.4443, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.5039, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.4439, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.0399, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.8264, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.6424, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.9802, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.3534, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.3107, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.7532, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.0589, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.3002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.3242, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.7196, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(139.5864, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(160.3585, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.7308, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.6228, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.2477, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.1956, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(61.6517, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.3548, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.0353, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.9235, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.6049, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.0760, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.1517, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(89.5234, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.22596754057428214, 'recall': 0.26500732064421667, 'f1': 0.2439353099730458, 'number': 683}, 'P': {'precision': 0.17117903930131004, 'recall': 0.23333333333333334, 'f1': 0.19748110831234258, 'number': 840}, 'overall_precision': 0.19373072970195274, 'overall_recall': 0.247537754432042, 'overall_f1': 0.21735370423753242, 'overall_accuracy': 0.6369416792620003}
			------------EPOCH 27---------------
Loss:  tensor(242.6488, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.3429, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(89.5493, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.8142, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.7727, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.0252, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.1664, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.9898, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.2923, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.4921, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.0635, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.5770, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.3447, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.6573, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.6358, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.8216, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.4257, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.4798, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.8879, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.0518, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.6465, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.6921, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.5127, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.9553, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(133.9022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(150.7037, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.0545, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.6845, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.3629, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.2254, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.9816, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.7005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.7474, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.6336, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.7304, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.2030, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.9723, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(73.2219, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.2248062015503876, 'recall': 0.2547584187408492, 'f1': 0.23884694577899795, 'number': 683}, 'P': {'precision': 0.1625659050966608, 'recall': 0.22023809523809523, 'f1': 0.1870576339737108, 'number': 840}, 'overall_precision': 0.18776150627615062, 'overall_recall': 0.23571897570584374, 'overall_f1': 0.20902474526928674, 'overall_accuracy': 0.6350654088705895}
			------------EPOCH 28---------------
Loss:  tensor(226.8764, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.0254, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(77.1619, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.5874, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.1655, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.7609, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.4493, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.8018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.0676, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.0181, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.6152, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.0333, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.8288, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.0826, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.2206, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5820, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.4451, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.0163, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.0736, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.1781, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.1814, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.9280, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.9858, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.8273, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(141.9700, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(157.0407, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.0471, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.2406, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5803, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.8692, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.7009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.8125, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.0527, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.0737, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.7228, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.8053, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.2266, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(69.4663, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.21125, 'recall': 0.24743777452415813, 'f1': 0.22791638570465272, 'number': 683}, 'P': {'precision': 0.1581196581196581, 'recall': 0.22023809523809523, 'f1': 0.18407960199004975, 'number': 840}, 'overall_precision': 0.17969543147208122, 'overall_recall': 0.2324359816152331, 'overall_f1': 0.20269109647867165, 'overall_accuracy': 0.6327200708813259}
			------------EPOCH 29---------------
Loss:  tensor(215.2958, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.5562, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(70.5654, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.0935, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.6771, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.8807, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.1185, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.9377, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.2431, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.4206, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.2935, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.8273, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.0027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.0546, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.4179, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.4352, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.3242, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.6829, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.9119, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.8172, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.9662, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.0135, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.3917, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.7510, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(134.9318, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(145.1171, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.8683, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.9869, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2423, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.1539, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.2252, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.5043, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.0251, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.7865, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.3141, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.1034, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.1151, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.1129, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.23252858958068615, 'recall': 0.2679355783308931, 'f1': 0.24897959183673468, 'number': 683}, 'P': {'precision': 0.16204869857262805, 'recall': 0.22976190476190475, 'f1': 0.19005416051206303, 'number': 840}, 'overall_precision': 0.19009100101112233, 'overall_recall': 0.2468811556139199, 'overall_f1': 0.2147957726363896, 'overall_accuracy': 0.6371501537499349}
			------------EPOCH 30---------------
Loss:  tensor(186.1487, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.2449, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(66.4382, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.5700, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.7210, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.8433, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.6277, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.4936, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.7059, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.2206, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.2177, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.1064, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.8403, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.7864, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.4212, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.1760, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.2403, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.2772, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.1200, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.5072, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.6257, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.5990, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.9271, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.6246, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(110.3548, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(117.7854, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.9457, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.0976, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.9099, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.9538, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.4950, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.0283, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.5300, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.3165, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.6833, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.3093, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.2574, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.8341, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.21311475409836064, 'recall': 0.24743777452415813, 'f1': 0.22899728997289973, 'number': 683}, 'P': {'precision': 0.16213847502191062, 'recall': 0.22023809523809523, 'f1': 0.1867743563856638, 'number': 840}, 'overall_precision': 0.18304033092037228, 'overall_recall': 0.2324359816152331, 'overall_f1': 0.2048018513161701, 'overall_accuracy': 0.639834262782092}
