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


		-------------RUN 1-----------
			------------EPOCH 1---------------
Loss:  tensor(1939.1482, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3612.3794, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1657.4824, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3203.6880, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2476.0828, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(597.6631, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1899.3481, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2563.1257, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1707.1229, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2263.3076, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2363.2769, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2277.9392, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2421.4067, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2238.1890, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2345.5918, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1543.1694, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2346.3237, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1542.7424, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3625.4709, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2710.6580, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2368.3442, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2197.5776, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1844.8909, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3007.8013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1954.5852, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1133.0686, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1529.3647, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1324.5277, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(739.5102, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1326.7363, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(769.4133, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1250.6123, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1988.9557, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1904.9352, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(987.6581, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1892.9392, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1866.3217, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1190.7600, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1904.8708, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2095.0520, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1400.0969, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2413.2222, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1570.9756, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2497.3196, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1450.3252, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1382.7257, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2734.5757, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1682.9695, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1724.5288, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(893.0114, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1453.5063, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1501.7708, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1815.6550, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1702.7502, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1022.3510, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1800.7610, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2618.3306, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2687.7759, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'number': 255}, 'P': {'precision': 0.0893970893970894, 'recall': 0.09267241379310345, 'f1': 0.091005291005291, 'number': 464}, 'overall_precision': 0.08301158301158301, 'overall_recall': 0.059805285118219746, 'overall_f1': 0.06952303961196442, 'overall_accuracy': 0.6409757152901725}
			------------EPOCH 2---------------
Loss:  tensor(1164.4958, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2104.9043, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(931.3378, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1851.4757, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1751.4287, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(388.1224, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1246.4180, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1631.1597, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1271.2799, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1633.0043, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1849.8203, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1983.2103, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1866.3384, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1536.2705, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1645.2441, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1185.4943, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1804.0769, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1299.9906, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3239.5896, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2313.1978, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2052.9375, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1951.9492, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1564.8282, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2548.2712, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1672.9874, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(966.7455, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1273.9044, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1077.2058, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(573.2633, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1074.4216, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(646.5005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1105.8564, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1814.6521, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1680.5066, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(864.3552, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1569.2979, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1448.9189, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(980.0753, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1513.3333, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1757.3698, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1159.6987, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2017.6211, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1192.5731, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2003.3330, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1260.6052, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1139.4734, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2247.8467, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1316.6526, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1393.6791, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(744.8761, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1235.7788, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1264.4026, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1570.3804, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1507.4806, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(895.1039, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1545.6802, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2286.6924, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2364.5093, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.11304347826086956, 'recall': 0.15294117647058825, 'f1': 0.13, 'number': 255}, 'P': {'precision': 0.13687150837988826, 'recall': 0.10560344827586207, 'f1': 0.11922141119221412, 'number': 464}, 'overall_precision': 0.1251778093883357, 'overall_recall': 0.12239221140472879, 'overall_f1': 0.12376933895921237, 'overall_accuracy': 0.6656931148250311}
			------------EPOCH 3---------------
Loss:  tensor(1032.3132, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1811.0359, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(768.4951, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1574.5673, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1510.3347, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(340.0842, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1038.8253, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1370.6047, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1045.0281, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1412.6650, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1547.6738, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1696.8561, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1560.6469, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1166.5422, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1265.1804, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(915.5675, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1448.7156, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1123.3110, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2932.2690, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1963.3276, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1708.1903, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1709.6605, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1291.2535, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2125.5913, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1371.8556, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(762.8256, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1012.1696, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(850.5612, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(437.0885, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(871.5304, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(472.8106, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(914.1314, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1492.9688, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1436.1307, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(652.8826, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1255.5337, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(917.6252, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(679.1786, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(969.2427, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1313.6102, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(978.2427, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1746.8950, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(932.3223, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1725.9084, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1082.1116, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(900.8395, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1859.0171, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1135.9003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1145.5414, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(618.9066, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1010.1357, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1061.8798, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1308.0886, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1253.7561, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(718.0314, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1231.8928, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1742.7117, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1840.2561, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.2149122807017544, 'recall': 0.19215686274509805, 'f1': 0.2028985507246377, 'number': 255}, 'P': {'precision': 0.1934782608695652, 'recall': 0.19181034482758622, 'f1': 0.19264069264069264, 'number': 464}, 'overall_precision': 0.2005813953488372, 'overall_recall': 0.19193324061196107, 'overall_f1': 0.19616204690831557, 'overall_accuracy': 0.728703553464222}
			------------EPOCH 4---------------
Loss:  tensor(796.7252, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1512.3264, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(610.9267, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1254.4792, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1241.9812, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(238.6855, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(868.4053, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1193.2716, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(853.9940, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1161.2441, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1244.8961, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1253.0396, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1256.6885, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(926.8018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1004.3180, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(681.2314, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1097.4478, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(902.1230, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2510.8228, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1500.0515, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1336.3738, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1379.6805, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(971.8579, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1812.3763, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1139.6204, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(588.7791, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(775.9038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(623.7034, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(299.4667, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(602.1382, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(347.6593, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(757.8012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1248.8250, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1263.4490, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(484.9347, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(939.1232, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(550.1873, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(492.7616, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(573.7347, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(987.0607, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(724.9979, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1400.2994, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(666.0734, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1415.1506, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(890.4761, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(703.1978, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1499.5176, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(912.5751, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(942.5244, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(485.9162, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(810.5293, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(812.4020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1086.3525, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1047.8097, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(602.3940, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(935.8021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1417.0103, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1522.6720, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.2074688796680498, 'recall': 0.19607843137254902, 'f1': 0.20161290322580644, 'number': 255}, 'P': {'precision': 0.16385135135135134, 'recall': 0.20905172413793102, 'f1': 0.18371212121212122, 'number': 464}, 'overall_precision': 0.17647058823529413, 'overall_recall': 0.20445062586926285, 'overall_f1': 0.18943298969072167, 'overall_accuracy': 0.7452539347720266}
			------------EPOCH 5---------------
Loss:  tensor(559.6501, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1267.6433, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(490.5243, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(990.1488, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(956.8445, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(169.0451, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(735.1243, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(949.0637, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(677.1237, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(902.9744, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(940.1216, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(927.9447, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1008.3914, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(646.6909, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(710.3528, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(477.2600, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(788.5863, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(720.9105, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1884.8942, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1055.0909, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1039.3013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1026.1937, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(744.9713, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1455.7711, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(983.4554, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(432.7216, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(572.9477, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(413.9512, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(225.5048, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(420.7896, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(326.5645, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(613.1687, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1331.9705, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1261.4059, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(384.2260, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(776.3070, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(408.6352, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(449.5625, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(386.8884, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(781.9645, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(503.7413, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1053.2288, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(461.6780, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1161.1792, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(808.3909, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(634.5453, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1440.2976, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(798.4045, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(944.9443, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(528.9177, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(808.2594, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(851.6127, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1033.8695, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(914.1994, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(358.0383, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(624.9014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1079.1882, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1140.6838, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.20673076923076922, 'recall': 0.16862745098039217, 'f1': 0.1857451403887689, 'number': 255}, 'P': {'precision': 0.2281303602058319, 'recall': 0.28663793103448276, 'f1': 0.2540592168099332, 'number': 464}, 'overall_precision': 0.2225031605562579, 'overall_recall': 0.24478442280945759, 'overall_f1': 0.23311258278145694, 'overall_accuracy': 0.7435772621558765}
			------------EPOCH 6---------------
Loss:  tensor(445.0760, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1236.8938, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(410.3922, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(779.7986, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(993.6307, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(195.4688, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(760.3041, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(957.6631, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(680.4340, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(931.0938, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1529.8433, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1684.4011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(827.3304, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(602.5796, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(551.4966, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(386.5576, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(588.6412, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(660.8914, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1546.1161, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(797.5699, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(829.0049, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(835.6305, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(667.5114, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1265.1089, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(954.3717, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(435.0148, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(604.5432, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(448.7034, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(210.3133, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(422.8948, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(299.6096, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(593.0372, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(978.5405, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1175.8564, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(416.8527, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(834.5250, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(286.4432, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(298.9997, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(333.4519, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(561.0013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(451.9922, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(899.1388, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(295.8456, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(941.4073, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(760.7461, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(480.8289, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1047.9873, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(531.1278, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(707.7552, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(365.7651, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(612.1412, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(589.9106, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(832.8098, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(755.7051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(297.5515, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(566.2074, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1057.3425, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1015.4571, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.170316301703163, 'recall': 0.27450980392156865, 'f1': 0.2102102102102102, 'number': 255}, 'P': {'precision': 0.2188235294117647, 'recall': 0.20043103448275862, 'f1': 0.20922384701912258, 'number': 464}, 'overall_precision': 0.19497607655502391, 'overall_recall': 0.2267037552155772, 'overall_f1': 0.20964630225080388, 'overall_accuracy': 0.7066363783871491}
			------------EPOCH 7---------------
Loss:  tensor(282.4333, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1131.7207, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(327.4053, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(526.4018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(660.8113, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(118.0776, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(687.9831, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(955.6029, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(542.4750, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(621.6683, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(740.0244, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(713.1881, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(678.0399, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(436.5257, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(420.5737, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(306.0242, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(477.0102, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(529.2661, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1753.0237, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(754.0642, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(732.9956, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(830.4990, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(575.0642, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1194.9109, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(623.7688, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(253.9567, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(404.8472, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(257.8637, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(110.4738, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(243.3594, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(211.0866, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(415.8903, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(744.3638, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(909.3853, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(323.2729, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(637.4993, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(214.8707, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(243.0279, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(237.3731, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(411.7351, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(314.8102, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(692.3901, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(297.8248, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(838.8522, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(498.0599, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(275.8825, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(668.2177, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(411.8108, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(530.7882, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(220.1484, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(409.1186, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(449.3482, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(658.5845, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(637.4642, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(220.2262, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(433.0242, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(748.2552, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(710.2200, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.18527918781725888, 'recall': 0.28627450980392155, 'f1': 0.22496147919876733, 'number': 255}, 'P': {'precision': 0.2052313883299799, 'recall': 0.21982758620689655, 'f1': 0.21227887617065555, 'number': 464}, 'overall_precision': 0.19640852974186307, 'overall_recall': 0.24339360222531292, 'overall_f1': 0.21739130434782608, 'overall_accuracy': 0.7356806749959436}
			------------EPOCH 8---------------
Loss:  tensor(224.3766, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(758.9836, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(217.3426, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(441.2458, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(590.5690, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(82.7318, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(390.1590, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(429.6353, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(320.9026, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(533.7028, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(665.8749, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(628.2578, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(657.9260, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(577.5927, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(294.0783, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(212.1593, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(308.7625, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(505.5526, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1153.5189, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(499.9825, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(417.6516, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(541.5869, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(485.1542, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(952.1711, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(698.1728, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(262.2889, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(408.7266, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(218.1881, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(86.2696, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(180.0122, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(205.5815, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(346.8818, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(728.5266, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1151.4423, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(221.9385, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(343.3410, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(198.7676, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(233.6790, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(201.7466, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(396.9871, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(221.0651, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(560.0084, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(301.1171, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(576.5891, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(449.6885, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(234.9779, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(550.9802, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(343.5658, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(479.7990, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(354.7841, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(518.7675, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(464.0653, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(813.1792, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(696.5564, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(228.1788, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(526.7430, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(837.1085, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(675.4107, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.18181818181818182, 'recall': 0.3686274509803922, 'f1': 0.24352331606217617, 'number': 255}, 'P': {'precision': 0.17771883289124668, 'recall': 0.14439655172413793, 'f1': 0.15933412604042807, 'number': 464}, 'overall_precision': 0.18008948545861297, 'overall_recall': 0.2239221140472879, 'overall_f1': 0.1996280223186609, 'overall_accuracy': 0.68667856563362}
			------------EPOCH 9---------------
Loss:  tensor(163.7883, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(634.2031, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(259.0801, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(348.2734, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(452.2247, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(91.8740, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(330.0285, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(371.0312, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(311.7377, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(406.7285, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(488.7404, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(428.0956, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(451.6218, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(366.0005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(182.7450, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(135.1106, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(291.4427, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(279.7395, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(992.3332, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(423.3868, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(365.4612, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(421.7903, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(332.9326, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(740.6533, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(370.9521, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(216.5005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(324.2403, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(162.5489, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(106.1500, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(152.7422, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(148.6338, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(286.6755, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(625.1184, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(781.3159, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(171.0483, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(308.6814, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(125.9091, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(132.2829, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(146.2270, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(286.8549, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(224.4766, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(532.0134, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(168.7986, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(556.2758, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(361.9384, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(185.7389, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(471.4347, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(273.0949, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(324.2254, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(126.4923, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(327.9640, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(327.4596, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(437.6498, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(449.8983, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(93.3641, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(252.5326, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(522.2128, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(557.0831, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.11584553928095873, 'recall': 0.3411764705882353, 'f1': 0.17296222664015906, 'number': 255}, 'P': {'precision': 0.06293706293706294, 'recall': 0.01939655172413793, 'f1': 0.02965403624382208, 'number': 464}, 'overall_precision': 0.10738255033557047, 'overall_recall': 0.13351877607788595, 'overall_f1': 0.11903285802851829, 'overall_accuracy': 0.5673643788198388}
			------------EPOCH 10---------------
Loss:  tensor(256.7891, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(810.5975, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(282.4618, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(422.2303, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(662.8540, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(142.1093, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(447.4289, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(531.5842, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(283.5288, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(581.7226, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(565.8472, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(462.9167, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(401.1390, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(271.6777, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(179.2554, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(204.7689, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(202.5248, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(198.2753, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(800.2469, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(291.4759, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(259.6481, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(365.0485, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(238.9219, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(602.0999, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(298.0496, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(104.8506, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(305.3005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(129.2176, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(82.5291, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(122.7319, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(98.9340, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(275.8225, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(892.3683, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1195.6604, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(253.8628, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(401.5245, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(284.8527, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(406.4677, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(243.0349, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(839.3024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(304.0287, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(944.8884, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(304.0806, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(647.2477, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(413.0602, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(213.0992, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(536.3479, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(202.5689, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(290.0051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(110.6075, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(181.1065, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(311.2646, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(318.5945, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(332.1749, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(69.0624, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(185.8743, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(344.1894, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(377.7582, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.10405405405405406, 'recall': 0.30196078431372547, 'f1': 0.15477386934673368, 'number': 255}, 'P': {'precision': 0.052083333333333336, 'recall': 0.010775862068965518, 'f1': 0.01785714285714286, 'number': 464}, 'overall_precision': 0.09808612440191387, 'overall_recall': 0.11404728789986092, 'overall_f1': 0.10546623794212219, 'overall_accuracy': 0.5211747525555736}
			------------EPOCH 11---------------
Loss:  tensor(154.6866, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(648.4831, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(200.8224, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(319.4369, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(737.6651, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(157.1473, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(411.2282, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(410.0590, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(429.3077, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(758.4694, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(868.5665, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(735.1698, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1620.3167, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1595.9724, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(956.5806, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(613.0094, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(538.4468, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(686.1544, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1474.6477, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(792.7990, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(396.7020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(547.5590, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(433.6857, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(760.7469, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(307.4973, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(99.6538, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(246.0257, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(130.3090, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(77.3552, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(115.0487, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(105.7769, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(221.8200, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(723.4243, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(851.8442, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(187.3994, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(401.4569, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(262.9517, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(296.7715, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(304.0754, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(578.8561, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(519.1674, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1193.8618, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(317.1428, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(917.8828, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(935.6937, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(921.7693, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1533.8082, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(578.1799, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1215.5109, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(625.8405, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(860.1827, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(964.8287, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1025.3583, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(729.5574, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(362.1388, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1274.4705, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(976.1205, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1097.1804, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.2857142857142857, 'recall': 0.24313725490196078, 'f1': 0.2627118644067796, 'number': 255}, 'P': {'precision': 0.18512396694214875, 'recall': 0.2413793103448276, 'f1': 0.20954162768942936, 'number': 464}, 'overall_precision': 0.2116788321167883, 'overall_recall': 0.24200278164116829, 'overall_f1': 0.22582738481505515, 'overall_accuracy': 0.7569365568716534}
			------------EPOCH 12---------------
Loss:  tensor(149.9528, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(610.0823, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(208.9370, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(340.2426, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(446.1929, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(91.1508, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(309.1140, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(333.6198, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(288.2705, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(492.6625, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(538.0448, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(489.8726, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(728.0953, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(581.8248, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(334.0518, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(230.6606, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(467.1682, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(547.1435, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1222.9524, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(701.7442, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(787.8541, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(864.2275, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(710.4482, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1146.9119, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(833.3635, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(276.1747, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(625.4108, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(332.4351, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(80.9404, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(270.3172, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(221.4902, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(439.9297, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(485.6332, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(698.9410, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(230.2988, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(426.9139, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(142.0442, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(173.3683, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(181.5995, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(330.6780, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(191.7982, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(547.6489, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(128.5426, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(592.7921, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(319.3891, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(153.7333, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(399.1210, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(346.2747, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(335.3693, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(134.9921, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(246.4082, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(320.1677, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(388.0759, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(343.3115, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(85.9704, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(248.5486, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(409.6250, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(362.1266, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.1946564885496183, 'recall': 0.2, 'f1': 0.19729206963249518, 'number': 255}, 'P': {'precision': 0.1806853582554517, 'recall': 0.25, 'f1': 0.2097649186256781, 'number': 464}, 'overall_precision': 0.1847345132743363, 'overall_recall': 0.23226703755215578, 'overall_f1': 0.20579174368453484, 'overall_accuracy': 0.7307588295743415}
			------------EPOCH 13---------------
Loss:  tensor(120.7174, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(664.7791, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(213.6562, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(267.6778, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(355.2182, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.0913, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(343.2842, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(362.9313, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(282.9224, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(440.4480, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(583.5057, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(422.1612, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(335.1863, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(259.0439, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(139.4799, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(153.4090, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(190.7702, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(199.6369, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(696.4121, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(342.3486, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(170.6607, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(269.7536, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(201.8356, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(477.2444, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(166.8125, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.6387, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(179.6912, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(98.6298, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.5374, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(85.2688, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(76.7533, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(170.7874, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(230.7515, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(376.3729, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(83.7963, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(126.5521, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(75.2345, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(94.0299, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(95.6105, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(171.8551, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(109.9697, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(272.1481, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.0384, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(256.6275, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(224.4694, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(95.4042, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(297.3070, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(136.2302, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(219.1965, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(65.3258, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(140.0825, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(192.1640, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(226.5520, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(240.7259, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.3263, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(128.1795, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(157.0598, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(183.3616, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.18489583333333334, 'recall': 0.2784313725490196, 'f1': 0.2222222222222222, 'number': 255}, 'P': {'precision': 0.19011406844106463, 'recall': 0.21551724137931033, 'f1': 0.202020202020202, 'number': 464}, 'overall_precision': 0.1879120879120879, 'overall_recall': 0.23783031988873435, 'overall_f1': 0.20994475138121546, 'overall_accuracy': 0.7166423278706258}
			------------EPOCH 14---------------
Loss:  tensor(49.3797, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(301.2159, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(69.4313, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(75.6131, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(273.6623, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.0370, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(165.7580, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(183.3814, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(110.0578, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(187.1884, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(300.5813, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(236.3958, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(211.5135, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(182.9338, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.5055, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.0358, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(102.7577, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(102.6403, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(482.0789, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(178.4052, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(104.3329, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(212.4724, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(152.4377, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(367.8057, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(123.4363, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.4197, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(175.6980, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(76.4996, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.0665, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(62.3592, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.1661, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(117.6991, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(148.0491, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(276.2905, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.3933, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(72.4954, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.2857, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(66.8489, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.4888, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(123.7564, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.3018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(212.4527, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.5138, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(160.3850, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(190.4547, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(75.3883, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(202.2765, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(84.4096, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(179.9409, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.5272, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(93.3246, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(196.0149, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(233.0637, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(249.2040, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.3596, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(140.2523, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(141.9525, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(174.8396, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.24444444444444444, 'recall': 0.25882352941176473, 'f1': 0.25142857142857145, 'number': 255}, 'P': {'precision': 0.17504332755632582, 'recall': 0.21767241379310345, 'f1': 0.1940441882804995, 'number': 464}, 'overall_precision': 0.19716646989374262, 'overall_recall': 0.23226703755215578, 'overall_f1': 0.2132822477650064, 'overall_accuracy': 0.7357347612093678}
			------------EPOCH 15---------------
Loss:  tensor(40.4771, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(237.3949, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.0064, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(64.6325, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(198.3473, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.2486, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(143.0522, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(149.6127, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(92.8735, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(137.9364, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(261.7390, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(214.6335, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(137.9411, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(129.2831, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.8514, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.4414, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(77.4680, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(68.6904, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(406.5410, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(134.6855, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(142.2194, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(207.3263, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(117.2015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(288.7140, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(77.5787, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.1595, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(133.5890, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.8694, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.9964, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.3872, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.2534, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(96.7306, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(128.5644, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(225.1868, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.6677, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(68.3695, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.9155, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(59.2906, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.6308, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(98.0242, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.5940, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(196.9463, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(71.6372, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(190.5103, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(117.1085, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.5929, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(155.5614, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(92.0403, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(109.8908, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.0365, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.6921, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(146.6607, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(162.3812, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(172.8369, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.7741, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(69.1600, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(179.9352, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(209.1151, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.17028985507246377, 'recall': 0.1843137254901961, 'f1': 0.1770244821092279, 'number': 255}, 'P': {'precision': 0.13931297709923665, 'recall': 0.15732758620689655, 'f1': 0.14777327935222673, 'number': 464}, 'overall_precision': 0.15, 'overall_recall': 0.16689847009735745, 'overall_f1': 0.15799868334430547, 'overall_accuracy': 0.68667856563362}
			------------EPOCH 16---------------
Loss:  tensor(41.5082, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(282.2234, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.6833, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(93.7473, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(271.0671, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.9000, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(124.0176, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(110.1661, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(92.0108, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(143.4308, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(262.6153, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(189.3383, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(142.7866, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(106.0267, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.9774, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.1764, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(61.3072, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.3400, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(320.3840, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(99.8065, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(83.1462, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(165.2726, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(86.0612, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(261.8195, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(66.0296, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.6131, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(93.8623, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.8058, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.5665, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.6361, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.2475, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(95.8610, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(145.7816, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(274.3463, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.1022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(89.2363, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.6042, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(73.5621, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.2959, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(147.1531, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(89.5856, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(348.3639, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.0303, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(291.9762, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(146.9181, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(124.4771, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(278.7230, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(160.3993, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(143.1955, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(70.1138, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.8289, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(238.4065, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(147.6016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(156.3866, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.8578, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(89.1907, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(102.4211, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(166.9386, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.18394648829431437, 'recall': 0.21568627450980393, 'f1': 0.19855595667870035, 'number': 255}, 'P': {'precision': 0.1960431654676259, 'recall': 0.2349137931034483, 'f1': 0.21372549019607842, 'number': 464}, 'overall_precision': 0.19181286549707602, 'overall_recall': 0.22809457579972184, 'overall_f1': 0.20838627700127066, 'overall_accuracy': 0.7416842446860296}
			------------EPOCH 17---------------
Loss:  tensor(22.6639, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(173.1817, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.6919, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.4919, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(168.9274, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.9253, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(77.6911, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.6249, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(104.7222, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(150.9263, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(239.7542, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(204.6631, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(214.2956, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(139.9690, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(150.8451, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(108.7382, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(173.5886, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(154.9637, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(572.8888, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(240.2234, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(178.6035, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(245.8773, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(127.1455, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(417.7160, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(67.7922, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.0247, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(91.0021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(62.4463, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.9594, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.7392, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.9784, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.5413, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(81.2409, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(164.4675, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.2011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.5631, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.0026, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.6110, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.6783, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(116.2891, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(120.4699, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(336.7502, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(108.5975, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(361.3032, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(219.2604, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(102.4588, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(273.2144, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(258.2565, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(219.8642, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(76.3842, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(155.1194, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(221.2281, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(153.9841, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(178.9926, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.6524, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(179.6821, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(126.1874, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(188.7144, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.30364372469635625, 'recall': 0.29411764705882354, 'f1': 0.29880478087649404, 'number': 255}, 'P': {'precision': 0.21682242990654205, 'recall': 0.25, 'f1': 0.23223223223223222, 'number': 464}, 'overall_precision': 0.2442455242966752, 'overall_recall': 0.26564673157162727, 'overall_f1': 0.2544970019986676, 'overall_accuracy': 0.7652117475255558}
			------------EPOCH 18---------------
Loss:  tensor(54.6585, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(256.8575, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.3594, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.1285, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(204.9229, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.8204, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(69.8273, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(112.3924, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(64.2716, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(116.5604, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(247.0478, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(173.5605, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(112.7682, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(87.8234, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.1050, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.3846, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(80.4948, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.2060, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(325.9103, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(94.9649, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(120.4518, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(181.2644, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(153.7965, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(465.6278, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(104.0614, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(105.8281, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(98.9656, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(156.7491, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.9596, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(77.4748, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(84.8821, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(164.3277, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(181.8522, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(313.2083, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.0699, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(81.5002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(62.3599, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.0347, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.9762, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(109.4884, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.8867, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(166.2133, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.5080, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(146.7132, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(71.1956, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.7995, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(131.8084, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(69.8149, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(86.7348, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.1943, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.8934, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(120.0260, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(116.3658, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(128.5026, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.1534, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(86.9002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(99.3298, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(196.7972, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.19458128078817735, 'recall': 0.30980392156862746, 'f1': 0.23903177004538578, 'number': 255}, 'P': {'precision': 0.18213660245183888, 'recall': 0.22413793103448276, 'f1': 0.20096618357487922, 'number': 464}, 'overall_precision': 0.18730808597748208, 'overall_recall': 0.2545201668984701, 'overall_f1': 0.21580188679245285, 'overall_accuracy': 0.7371950889718211}
			------------EPOCH 19---------------
Loss:  tensor(45.2453, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(185.0207, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.9631, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.1861, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(197.9867, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.9782, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(69.9740, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.7190, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.5618, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(176.0871, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(457.8696, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(485.0031, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(253.8012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(193.2904, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(89.8977, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.9893, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(105.8336, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(102.0388, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(286.2831, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(88.7007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.4858, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(130.5095, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(88.3204, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(280.5983, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(69.5422, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.7547, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.9058, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.8289, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.0370, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.8565, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.7974, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.4532, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(64.2962, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(148.7362, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.5107, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.6287, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.3423, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.6832, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.8698, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(89.4319, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(62.7150, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(192.6158, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.8143, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(153.7195, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(161.5047, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(70.8509, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(237.3528, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(229.1891, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(183.6986, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.7431, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(87.0800, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(198.6263, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(138.7163, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(153.9535, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.0595, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.5053, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.5009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(114.3533, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.2450592885375494, 'recall': 0.24313725490196078, 'f1': 0.24409448818897636, 'number': 255}, 'P': {'precision': 0.2066420664206642, 'recall': 0.2413793103448276, 'f1': 0.22266401590457258, 'number': 464}, 'overall_precision': 0.2188679245283019, 'overall_recall': 0.24200278164116829, 'overall_f1': 0.2298546895640687, 'overall_accuracy': 0.7555844015360484}
			------------EPOCH 20---------------
Loss:  tensor(14.0631, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(159.2772, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.8778, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.5009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(100.2621, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.2990, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.4737, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.2613, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.3468, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(59.8853, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(252.7298, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(171.2064, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(94.4794, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(81.6865, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.7578, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.6090, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.0981, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.3096, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(246.4121, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(66.7635, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.9678, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(105.0551, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(90.9065, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(247.9338, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(86.8391, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.0447, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.7738, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.0751, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.6373, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.6025, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.7883, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.3320, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(82.8175, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(177.8381, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.5869, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.2541, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.6556, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.8620, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.2021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.1304, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.1344, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(126.1015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.7213, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(64.4167, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(67.3305, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(59.2361, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(184.9108, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(66.6349, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(77.4124, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.6865, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.7836, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(138.2938, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(104.6663, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(109.7816, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.1778, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.6387, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.6523, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(93.2522, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.19101123595505617, 'recall': 0.26666666666666666, 'f1': 0.2225859247135843, 'number': 255}, 'P': {'precision': 0.17543859649122806, 'recall': 0.23706896551724138, 'f1': 0.2016498625114574, 'number': 464}, 'overall_precision': 0.18107833163784334, 'overall_recall': 0.24756606397774686, 'overall_f1': 0.20916568742655697, 'overall_accuracy': 0.7454161934122993}
			------------EPOCH 21---------------
Loss:  tensor(10.0588, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(123.1510, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.3728, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.9496, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(92.3587, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.1530, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.3243, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.6930, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.0487, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(65.6891, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(196.7420, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(100.0256, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(92.9643, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(79.1638, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.2530, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.4002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.4213, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.4945, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(188.4440, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.3685, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.1668, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(96.4253, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.9739, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(151.7236, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.9173, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.2951, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.9733, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.4240, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.0263, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.6695, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.6820, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.9177, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.4137, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(102.4925, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.1264, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.2733, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.1892, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.5525, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.9256, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.8758, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.0696, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(95.8392, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.1189, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.9097, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(79.6478, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(74.0999, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(182.1155, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.0523, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.1212, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.7251, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.1826, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(103.1256, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(74.1359, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(76.6762, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.5953, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.9675, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.0417, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(73.8939, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.23711340206185566, 'recall': 0.27058823529411763, 'f1': 0.25274725274725274, 'number': 255}, 'P': {'precision': 0.20809248554913296, 'recall': 0.23275862068965517, 'f1': 0.219735503560529, 'number': 464}, 'overall_precision': 0.21851851851851853, 'overall_recall': 0.24617524339360222, 'overall_f1': 0.2315238718116416, 'overall_accuracy': 0.7516901941695062}
			------------EPOCH 22---------------
Loss:  tensor(10.0286, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(105.3693, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.6199, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.1810, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(62.6277, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.9955, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.8344, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.2302, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.7411, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.0848, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(143.3106, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(96.9433, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(65.8613, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.2467, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.2519, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.1505, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.2385, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.4382, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(156.5676, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.9761, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.4451, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(70.4114, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.3738, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(125.9501, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.9563, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.5736, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.6318, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.1011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.0061, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.6036, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.5130, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.6871, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.1992, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(76.1826, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.9240, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.9504, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.5431, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.5073, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.8791, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.7605, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.0513, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(70.3666, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.2663, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.3283, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.5571, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(108.3433, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.2754, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.4845, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.9890, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.4442, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(68.6347, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(62.6112, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(66.1836, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.5283, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.9456, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.0957, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.9574, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.22522522522522523, 'recall': 0.29411764705882354, 'f1': 0.25510204081632654, 'number': 255}, 'P': {'precision': 0.1863468634686347, 'recall': 0.21767241379310345, 'f1': 0.20079522862823063, 'number': 464}, 'overall_precision': 0.20114285714285715, 'overall_recall': 0.24478442280945759, 'overall_f1': 0.22082810539523215, 'overall_accuracy': 0.7420087619665747}
			------------EPOCH 23---------------
Loss:  tensor(5.7684, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(78.2811, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.2344, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.2330, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.1966, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.4242, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.5073, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.8775, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.6746, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.2792, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(117.8676, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(86.6491, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.2316, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.5168, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.5238, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.2968, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.9962, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.5967, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(140.0258, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.5475, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.0957, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.4621, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.7202, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(108.6920, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.8781, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.5845, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.8717, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.1712, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.9320, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.1611, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.1595, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.3717, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.3727, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.0446, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.9356, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.8377, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.9278, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.2149, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.5062, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.9707, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.2688, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.4814, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.9547, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.6988, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.5230, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.1638, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(69.6848, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.1762, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.1141, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.6190, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.4972, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.1552, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.4276, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.3356, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.5121, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.6865, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.1600, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.9341, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.22121212121212122, 'recall': 0.28627450980392155, 'f1': 0.24957264957264955, 'number': 255}, 'P': {'precision': 0.17830882352941177, 'recall': 0.20905172413793102, 'f1': 0.19246031746031744, 'number': 464}, 'overall_precision': 0.1945080091533181, 'overall_recall': 0.2364394993045897, 'overall_f1': 0.21343377275580666, 'overall_accuracy': 0.7423332792471199}
			------------EPOCH 24---------------
Loss:  tensor(4.8839, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(65.5172, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.2949, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.2907, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.9660, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.2677, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.5283, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.2546, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.7155, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.6798, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(108.8826, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(75.0977, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.8596, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.1654, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.8088, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.2622, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.7450, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.8238, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(124.7979, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.3804, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.6777, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.1539, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.9205, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(87.5928, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.5225, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.5349, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.8731, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.2405, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6353, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.1841, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.0762, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.0462, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.9119, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.5288, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.3799, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.5733, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.1268, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.1063, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.3791, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.9043, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.2775, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.4142, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.8370, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.2693, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.7814, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.5572, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(71.5959, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.5182, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.4239, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.5472, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.0152, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.8346, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.5434, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.4010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.5871, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.4767, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.6897, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.9913, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.24916943521594684, 'recall': 0.29411764705882354, 'f1': 0.2697841726618705, 'number': 255}, 'P': {'precision': 0.17830882352941177, 'recall': 0.20905172413793102, 'f1': 0.19246031746031744, 'number': 464}, 'overall_precision': 0.20355029585798817, 'overall_recall': 0.23922114047287898, 'overall_f1': 0.21994884910485935, 'overall_accuracy': 0.7446049002109363}
			------------EPOCH 25---------------
Loss:  tensor(3.8958, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(67.5609, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.3823, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.4563, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.3817, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.6697, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.7942, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.8028, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.8110, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.2546, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(104.0762, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(79.4239, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.9311, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.4817, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.1713, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.0284, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.6661, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.5930, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(112.3159, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(65.6232, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.2021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.1651, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.6762, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(75.5049, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.7677, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.7648, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.8760, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.1162, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2714, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.0479, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.4126, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.5046, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.3424, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.2495, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.7914, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.1130, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.2605, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.4255, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.5743, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.7675, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.2398, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.8489, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5048, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.3415, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.5818, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.2507, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(98.5619, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.6325, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.0411, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.5250, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.8611, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(71.7115, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.9996, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.4302, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.7109, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.1567, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.4558, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.6671, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.22126436781609196, 'recall': 0.30196078431372547, 'f1': 0.2553897180762852, 'number': 255}, 'P': {'precision': 0.19498069498069498, 'recall': 0.21767241379310345, 'f1': 0.20570264765784113, 'number': 464}, 'overall_precision': 0.20554272517321015, 'overall_recall': 0.24756606397774686, 'overall_f1': 0.2246056782334385, 'overall_accuracy': 0.7403320893504246}
			------------EPOCH 26---------------
Loss:  tensor(3.3175, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(76.7257, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.7096, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.6269, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.1083, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.0198, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.2793, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.0453, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.7704, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.0166, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(100.6631, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.0536, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(80.4103, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(76.8651, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.3876, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.3177, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.8966, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.9604, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(93.1341, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.9421, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.0909, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.5452, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.2019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(70.2576, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.6228, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.3612, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.3410, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.4249, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1378, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.1995, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.7388, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.7264, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.6428, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.9095, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.0502, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.1865, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.9686, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.5757, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.8359, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.2076, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.4707, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.7919, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.9359, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.5855, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.8996, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.3183, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.2225, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.7911, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.6500, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.0157, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.8354, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.2116, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(62.4414, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(62.1032, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.1617, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.4079, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.2537, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.0061, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.24666666666666667, 'recall': 0.2901960784313726, 'f1': 0.2666666666666667, 'number': 255}, 'P': {'precision': 0.17192982456140352, 'recall': 0.21120689655172414, 'f1': 0.18955512572533847, 'number': 464}, 'overall_precision': 0.19770114942528735, 'overall_recall': 0.23922114047287898, 'overall_f1': 0.21648835745752046, 'overall_accuracy': 0.7451457623451783}
			------------EPOCH 27---------------
Loss:  tensor(2.8528, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(62.9795, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.7162, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.2698, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.2686, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.4794, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.6565, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.5572, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.9846, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.9402, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(90.3079, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(86.3997, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(72.8253, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(104.6737, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.1277, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.6662, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.4171, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.6307, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(86.3713, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.7917, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.6183, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.4358, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.9663, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(69.1281, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.7939, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.0664, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.1340, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9000, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.9554, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.4353, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.1986, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.1956, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.7771, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.8832, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.3800, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.8750, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.8507, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.7131, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.3033, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.6913, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.2291, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.7223, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.2855, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.0807, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.0405, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.2441, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.6141, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.0953, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.2870, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.3705, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.5658, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.6202, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.4472, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.3689, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.4470, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.6331, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.2591, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.2206703910614525, 'recall': 0.30980392156862746, 'f1': 0.2577487765089723, 'number': 255}, 'P': {'precision': 0.19560878243512975, 'recall': 0.21120689655172414, 'f1': 0.20310880829015543, 'number': 464}, 'overall_precision': 0.2060535506402794, 'overall_recall': 0.24617524339360222, 'overall_f1': 0.22433460076045628, 'overall_accuracy': 0.7401157444967278}
			------------EPOCH 28---------------
Loss:  tensor(2.6728, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(62.2727, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.4722, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.7747, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.8887, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2339, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.6832, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.2765, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.3538, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.0122, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(84.5736, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.3703, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(71.5979, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.4572, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.6401, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.3678, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.5820, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.4771, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(94.0128, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.7162, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.0229, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.3027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.2238, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(66.9531, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.3074, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.8045, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.0572, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.7833, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7900, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.3799, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.0244, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.3651, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.9566, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.6864, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2662, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.9038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.9171, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2841, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.4903, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.7682, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.3604, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.9794, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6330, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.0521, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.0512, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.9864, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.7152, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.6138, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.1356, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.7790, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.7183, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.8300, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.9234, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.4379, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.7518, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.2902, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.0764, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.3137, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.2550335570469799, 'recall': 0.2980392156862745, 'f1': 0.27486437613019893, 'number': 255}, 'P': {'precision': 0.1736111111111111, 'recall': 0.21551724137931033, 'f1': 0.1923076923076923, 'number': 464}, 'overall_precision': 0.20137299771167047, 'overall_recall': 0.24478442280945759, 'overall_f1': 0.22096672944130571, 'overall_accuracy': 0.745091676131754}
			------------EPOCH 29---------------
Loss:  tensor(2.2355, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.8977, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.8839, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5779, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.5457, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.7634, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.0311, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.7512, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.0928, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.4493, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(75.6343, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(81.8357, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.4428, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.6228, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.4262, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.9793, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.2345, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.8494, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(82.5646, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.3418, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.7310, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.0992, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.4094, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.9188, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.1736, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6032, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.1787, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5730, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7188, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.1294, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.7303, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.6718, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.9495, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.7537, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.0250, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5935, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.0450, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.8859, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.9366, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.7784, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.0250, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.0121, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2713, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.7623, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.9124, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.0687, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.5056, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.1787, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.5465, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.4268, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.3792, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.5243, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.8930, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.5631, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2134, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.3673, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.9875, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.2653, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.2215909090909091, 'recall': 0.3058823529411765, 'f1': 0.257001647446458, 'number': 255}, 'P': {'precision': 0.19883040935672514, 'recall': 0.21982758620689655, 'f1': 0.2088024564994882, 'number': 464}, 'overall_precision': 0.20809248554913296, 'overall_recall': 0.25034770514603616, 'overall_f1': 0.22727272727272727, 'overall_accuracy': 0.7391421926550922}
			------------EPOCH 30---------------
Loss:  tensor(2.0965, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.1709, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5855, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.1823, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.8272, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6489, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.6307, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.6289, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.4261, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.0754, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(82.0885, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.5509, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.9476, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.6341, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.1131, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.8008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.2961, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.9074, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(70.2772, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.9303, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.9572, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.1699, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.7432, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.8953, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.9923, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4142, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.1664, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4097, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6378, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.8926, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5310, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.1521, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.0804, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.9068, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.7637, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.3389, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.3733, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6559, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.4439, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.0420, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.7801, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.8746, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1828, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.3257, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.5906, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.7418, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.1328, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.7323, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.9012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.1834, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.0166, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.8500, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.6101, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.0872, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.7821, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.9913, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.2878, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.8559, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.2508250825082508, 'recall': 0.2980392156862745, 'f1': 0.27240143369175623, 'number': 255}, 'P': {'precision': 0.18197879858657243, 'recall': 0.22198275862068967, 'f1': 0.19999999999999998, 'number': 464}, 'overall_precision': 0.20598388952819333, 'overall_recall': 0.24895688456189152, 'overall_f1': 0.22544080604534006, 'overall_accuracy': 0.7482286765103575}


		-------------RUN 2-----------
			------------EPOCH 1---------------
Loss:  tensor(3085.4756, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2376.3516, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3856.3447, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2015.6921, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2073.3706, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2817.6299, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1928.7822, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3159.2349, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2665.7461, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3093.8574, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2749.3584, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1401.7750, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(826.3069, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1852.0601, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2286.9326, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1536.5835, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1639.2889, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1744.0374, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1333.5573, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1546.5684, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1502.2063, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1160.0051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1249.7968, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1432.6008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2751.7034, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2832.9580, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(774.3806, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1671.2891, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2620.0388, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2049.6384, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1696.9429, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3236.3867, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2426.8047, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1010.9315, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(825.6711, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1471.0547, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1706.3848, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2489.5198, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1391.9120, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1058.7793, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2051.5164, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1238.0596, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2213.3198, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1120.9658, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1588.6343, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1299.1621, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1696.1046, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2140.0725, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1404.3756, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2006.4722, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1727.4822, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2029.5223, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1049.7379, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1531.8024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2263.2998, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1215.9230, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1775.2314, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.06870229007633588, 'recall': 0.02586206896551724, 'f1': 0.03757828810020877, 'number': 348}, 'P': {'precision': 0.15471698113207547, 'recall': 0.27986348122866894, 'f1': 0.19927095990279461, 'number': 293}, 'overall_precision': 0.13767019667170954, 'overall_recall': 0.1419656786271451, 'overall_f1': 0.13978494623655915, 'overall_accuracy': 0.5575832173819972}
			------------EPOCH 2---------------
Loss:  tensor(1606.2715, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1179.4009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2090.6904, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1016.9219, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1326.8696, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2301.4731, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1282.8845, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1915.8624, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1734.4417, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2222.8811, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1909.5818, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1107.6514, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(589.6765, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1296.2358, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1505.1726, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1128.5173, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1169.5918, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1385.5072, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1062.3092, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1211.5436, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1208.2744, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(909.2245, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(998.6647, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1100.6917, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2255.2383, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2316.1724, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(582.7507, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1270.9685, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2162.6707, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1642.1272, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1351.5184, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2861.4170, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2068.7964, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(838.7094, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(658.6285, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1196.2145, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1322.4608, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1823.8333, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1110.5088, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(857.9577, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1692.0339, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1033., device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1786.9194, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(947.2551, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1214.2756, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1068.4436, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1303.2825, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1859.5687, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1197.1926, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1680.5081, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1372.8770, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1739.1870, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(849.9802, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1226.4517, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1581.9302, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(899.2302, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1125.0437, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.0896358543417367, 'recall': 0.09195402298850575, 'f1': 0.09078014184397162, 'number': 348}, 'P': {'precision': 0.05502392344497608, 'recall': 0.07849829351535836, 'f1': 0.06469760900140646, 'number': 293}, 'overall_precision': 0.07096774193548387, 'overall_recall': 0.08580343213728549, 'overall_f1': 0.07768361581920905, 'overall_accuracy': 0.6185379428449106}
			------------EPOCH 3---------------
Loss:  tensor(1436.4244, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(875.1072, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1674.2074, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(926.6116, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1074.3538, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1687.7686, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1094.9174, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1574.0670, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1365.8564, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1794.1141, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1494.6937, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(959.2457, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(466.4596, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1085.0994, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1036.5154, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(924.0933, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(886.0204, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1187.5093, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1012.6993, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1015.6919, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(986.1921, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(732.1526, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(813.8981, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1058.5875, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1853.7209, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1895.4216, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(399.7681, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(999.8342, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1700.6353, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1273.5233, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(925.1372, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2479.6438, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1633.8250, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(695.1161, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(450.7003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(991.3660, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1210.0923, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1556.1555, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(895.4149, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(613.5187, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1336.8101, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(886.5509, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1477.3120, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(719.7830, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(957.9946, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(898.2338, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1026.0400, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1535.6826, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(961.3914, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1216.6448, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(980.9995, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1299.6086, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(757.2050, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1040.9578, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(997.8591, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(643.7657, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(576.0627, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.11814345991561181, 'recall': 0.16091954022988506, 'f1': 0.13625304136253039, 'number': 348}, 'P': {'precision': 0.20967741935483872, 'recall': 0.26621160409556316, 'f1': 0.23458646616541357, 'number': 293}, 'overall_precision': 0.15839243498817968, 'overall_recall': 0.20904836193447737, 'overall_f1': 0.18022864828513785, 'overall_accuracy': 0.6350208712405009}
			------------EPOCH 4---------------
Loss:  tensor(1048.0780, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(521.4983, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1301.2288, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(689.8469, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(878.7623, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1376.4673, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(901.1603, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1251.6292, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1093.2603, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1355.0598, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1187.8685, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(742.0197, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(340.5538, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(872.3126, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(778.3478, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(807.5911, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(755.1572, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(917.6191, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(716.7183, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(760.6675, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(765.6240, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(526.2612, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(571.4973, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(601.1876, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1407.8156, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1466.9199, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(284.3487, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(818.9543, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1384.1050, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1032.2007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(656.6644, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2054.5688, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1101.3291, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(581.2026, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(318.1618, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(797.5529, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(973.3046, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1311.3729, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(605.9936, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(431.9653, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1006.9420, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(753.3810, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1253.6202, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(534.5085, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(754.5232, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(727.8064, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(807.6619, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1317.8032, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(650.0946, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(937.2573, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(701.5144, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(863.5137, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(673.8201, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(858.5002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(673.4189, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(520.3386, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(292.8004, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.14814814814814814, 'recall': 0.1724137931034483, 'f1': 0.1593625498007968, 'number': 348}, 'P': {'precision': 0.20995670995670995, 'recall': 0.3310580204778157, 'f1': 0.25695364238410595, 'number': 293}, 'overall_precision': 0.18108419838523646, 'overall_recall': 0.24492979719188768, 'overall_f1': 0.2082228116710875, 'overall_accuracy': 0.6304720111313282}
			------------EPOCH 5---------------
Loss:  tensor(839.0093, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(333.4882, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(955.5407, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(512.3424, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(703.7766, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1106.3401, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(737.7738, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1007.1320, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(912.9423, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1098.2349, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(956.3083, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(640.1434, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(257.1170, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(819.7972, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(686.5925, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(550.1122, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(548.6456, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(786.0162, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(592.6988, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(531.9939, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(587.0818, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(345.8205, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(347.6391, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(381.6750, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1299.7432, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1301.5245, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(268.0093, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(780.0906, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1089.5762, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(805.2030, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(485.0093, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2014.1489, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(747.2780, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(386.1369, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(231.4147, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(646.2396, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(776.5050, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1111.6033, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(412.1247, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(301.9614, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1067.8441, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(742.2938, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1208.3306, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(499.4277, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(826.9508, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(659.2878, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(781.2275, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1449.9480, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(743.7964, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1371.9154, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(801.4894, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1187.2239, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(544.2723, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(650.8167, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(395.3871, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(295.8175, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(166.2748, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3262411347517731, 'recall': 0.13218390804597702, 'f1': 0.18813905930470348, 'number': 348}, 'P': {'precision': 0.19906687402799378, 'recall': 0.43686006825938567, 'f1': 0.27350427350427353, 'number': 293}, 'overall_precision': 0.22193877551020408, 'overall_recall': 0.2714508580343214, 'overall_f1': 0.2442105263157895, 'overall_accuracy': 0.5963823183131757}
			------------EPOCH 6---------------
Loss:  tensor(977.2384, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(332.4989, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(934.6997, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(488.5725, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(925.0604, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1372.8501, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(804.7972, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1364.2017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1052.9376, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1542.4519, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1530.8887, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(742.2039, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(267.4903, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(760.1960, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(640.6221, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(631.1448, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(439.3084, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(789.9634, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(531.2820, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(407.4203, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(716.2748, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(306.7881, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(335.6354, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(309.1974, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1412.3906, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1482.7396, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(210.1845, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(693.2520, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1085.2340, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(899.2189, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(803.3993, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1816.5349, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(949.3000, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(312.5509, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(215.3189, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(631.3845, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(732.3778, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1056.7845, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(376.7518, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(265.7769, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(811.4443, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(649.1289, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1055.8268, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(394.4287, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(618.1101, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(553.0712, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(633.9872, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1129.8708, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(540.2279, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(698.2202, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(552.4785, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(708.6728, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(462.6127, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(528.5022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(332.3207, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(282.8364, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(132.0779, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.22945205479452055, 'recall': 0.1925287356321839, 'f1': 0.209375, 'number': 348}, 'P': {'precision': 0.15102974828375287, 'recall': 0.22525597269624573, 'f1': 0.1808219178082192, 'number': 293}, 'overall_precision': 0.18244170096021947, 'overall_recall': 0.20748829953198128, 'overall_f1': 0.19416058394160585, 'overall_accuracy': 0.6642406079417746}
			------------EPOCH 7---------------
Loss:  tensor(825.7068, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(265.4043, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(736.5640, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(436.7571, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(628.9563, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1374.7887, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(468.7819, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1076.5852, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(699.7775, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(699.5358, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(614.4594, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(448.1355, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(163.8962, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(552.1724, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(513.7847, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(428.2586, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(383.0121, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(740.6842, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(584.6038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(427.1724, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(567.3691, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(299.9707, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(320.2856, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(575.2960, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1131.0040, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1200.6089, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(206.3517, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(461.6828, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(782.5049, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(646.7324, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(470.8647, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1459.4960, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(727.9361, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(239.3013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(222.6937, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(642.6274, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(666.1111, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(889.6221, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(369.6370, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(259.9231, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(564.7620, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(415.1738, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(978.1317, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(236.3871, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(443.3996, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(434.8586, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(424.0212, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1107.6901, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(441.7499, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(510.4207, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(352.9870, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(491.9413, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(432.3768, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(400.7502, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(287.3178, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(227.7182, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(127.5873, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.16444444444444445, 'recall': 0.31896551724137934, 'f1': 0.21700879765395897, 'number': 348}, 'P': {'precision': 0.20588235294117646, 'recall': 0.09556313993174062, 'f1': 0.13053613053613053, 'number': 293}, 'overall_precision': 0.17139334155363747, 'overall_recall': 0.21684867394695787, 'overall_f1': 0.19146005509641872, 'overall_accuracy': 0.6026436904634486}
			------------EPOCH 8---------------
Loss:  tensor(969.4979, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(283.7133, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(917.7857, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(596.4646, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(726.4395, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(720.9425, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(601.0865, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(997.1118, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(618.3018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(617.2758, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(565.6756, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(444.3521, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(128.3290, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(418.5908, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(351.1035, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(302.6363, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(248.3359, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(494.9128, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(420.0776, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(257.1955, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(353.5505, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(209.5176, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(187.9220, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(236.0037, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1041.4263, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(928.1859, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(189.7353, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(458.6354, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(829.0865, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(608.2482, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(312.5393, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1526.1509, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(718.8942, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(191.6569, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(231.9705, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(622.2316, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(519.0599, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(711.5052, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(234.1151, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(194.5317, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(391.1151, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(374.3696, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(852.0082, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(136.6726, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(363.9028, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(324.7161, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(389.9901, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(994.3695, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(376.7434, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(383.8050, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(315.6982, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(425.6251, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(307.9349, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(283.1639, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(204.7341, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(199.2970, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(87.1639, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.1475195822454308, 'recall': 0.32471264367816094, 'f1': 0.2028725314183124, 'number': 348}, 'P': {'precision': 0.047619047619047616, 'recall': 0.023890784982935155, 'f1': 0.03181818181818182, 'number': 293}, 'overall_precision': 0.13143483023001096, 'overall_recall': 0.187207488299532, 'overall_f1': 0.15444015444015446, 'overall_accuracy': 0.6090656106175747}
			------------EPOCH 9---------------
Loss:  tensor(604.7006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(218.3758, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(687.2209, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(469.2161, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(630.3969, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(700.1311, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(518.0472, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(937.6511, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(767.1851, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(821.2013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(632.4780, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(550.1780, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(179.1119, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(539.1088, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(695.5682, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(372.8430, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(443.5023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(662.8744, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(378.1778, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(240.1108, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(303.1240, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(119.8833, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(148.3263, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(178.1873, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(510.8292, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(493.1605, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(105.9152, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(246.6937, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(408.5039, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(323.8730, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(233.7355, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(863.0386, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(447.6880, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(153.1476, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(198.1543, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(451.9474, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(589.2356, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(696.0541, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(352.3122, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(235.4247, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(559.1081, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(518.5869, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1281.1854, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(314.1194, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(725.3321, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(498.2879, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(472.0376, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1399.9290, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(507.4248, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(493.2263, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(282.4096, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(462.2447, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(351.6438, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(372.0927, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(228.0292, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(240.8753, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.1473, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.23209169054441262, 'recall': 0.23275862068965517, 'f1': 0.23242467718794835, 'number': 348}, 'P': {'precision': 0.1889763779527559, 'recall': 0.24573378839590443, 'f1': 0.2136498516320475, 'number': 293}, 'overall_precision': 0.2095890410958904, 'overall_recall': 0.23868954758190328, 'overall_f1': 0.22319474835886216, 'overall_accuracy': 0.6656320239751686}
			------------EPOCH 10---------------
Loss:  tensor(427.2530, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(128.1545, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(400.5449, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(264.3853, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(277.3250, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(567.4232, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(225.2083, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(444.1284, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(466.1117, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(387.3024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(317.3652, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(332.3606, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(122.3535, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(406.9642, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(465.7149, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(309.4108, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(484.5905, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(681.1407, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(345.9820, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(259.1435, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(519.9496, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(203.9413, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(290.6646, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(263.2768, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1180.5670, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(996.3052, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(159.1222, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(513.2223, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(647.4498, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(488.3198, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(439.8487, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(950.6957, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(450.6164, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(164.5521, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(81.8270, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(411.1657, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(404.5748, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(596.6619, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(116.2127, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(82.2988, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(303.0790, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(288.7327, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(580.7014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(84.5934, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(306.3098, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(275.9008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(329.9570, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(625.7041, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(349.9030, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(326.4756, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(202.1642, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(380.5350, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(367.4664, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(344.2888, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(284.0865, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(294.0830, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(94.1307, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.304635761589404, 'recall': 0.13218390804597702, 'f1': 0.18436873747494992, 'number': 348}, 'P': {'precision': 0.20065789473684212, 'recall': 0.41638225255972694, 'f1': 0.27081021087680357, 'number': 293}, 'overall_precision': 0.22134387351778656, 'overall_recall': 0.2620904836193448, 'overall_f1': 0.24, 'overall_accuracy': 0.6205715508937172}
			------------EPOCH 11---------------
Loss:  tensor(814.5786, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(259.1865, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(616.4623, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(276.0311, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(417.3683, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(856.5549, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(329.6057, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(733.6792, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(520.1993, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(475.1976, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(411.7977, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(368.4751, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(124.6623, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(451.6025, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(290.6490, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(262.9251, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(184.5227, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(388.9810, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(300.0797, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(170.1668, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(187.4199, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(80.9504, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(96.9771, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(113.8069, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(354.0958, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(350.9250, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(86.8040, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(197.5992, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(356.1748, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(235.7757, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(230.6483, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(613.1567, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(334.1987, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(99.3226, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(65.2953, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(345.2515, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(374.0242, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(603.5941, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(112.9609, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(92.4005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(310.2817, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(281.9281, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(577.5165, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(88.3983, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(236.3389, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(230.7545, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(245.9663, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(604.9124, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(276.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(215.0640, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(196.5665, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(300.4265, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(201.4299, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(249.2681, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(144.6682, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(86.3313, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.0836, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.21818181818181817, 'recall': 0.2413793103448276, 'f1': 0.2291950886766712, 'number': 348}, 'P': {'precision': 0.15984405458089668, 'recall': 0.27986348122866894, 'f1': 0.20347394540942929, 'number': 293}, 'overall_precision': 0.18485523385300667, 'overall_recall': 0.2589703588143526, 'overall_f1': 0.21572449642625083, 'overall_accuracy': 0.6631167719148026}
			------------EPOCH 12---------------
Loss:  tensor(277.3674, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(77.3818, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(318.4066, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(201.2888, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(183.1746, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(611.4971, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(162.7154, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(307.1562, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(245.1315, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(252.2871, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(192.9097, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(211.7633, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.0910, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(303.4192, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(176.0811, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(171.6802, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(125.2251, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(325.1360, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(194.4541, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(95.0553, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(182.5707, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(67.2353, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(81.3717, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(100.0191, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(324.9495, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(301.6252, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(85.8144, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(157.1707, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(300.7797, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(197.4222, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(173.2692, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(637.7025, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(248.3896, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.1514, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.2024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(286.4109, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(252.1259, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(423.2408, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(59.0828, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.5196, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(186.4284, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(167.6185, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(406.6411, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.5919, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(164.2928, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(161.8691, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(204.6405, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(384.4628, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(189.5059, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(116.9170, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(118.4701, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(198.5369, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(151.3575, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(169.4620, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(127.0903, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(74.4699, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.7360, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.2509090909090909, 'recall': 0.19827586206896552, 'f1': 0.2215088282504013, 'number': 348}, 'P': {'precision': 0.14257425742574256, 'recall': 0.24573378839590443, 'f1': 0.18045112781954883, 'number': 293}, 'overall_precision': 0.18076923076923077, 'overall_recall': 0.21996879875195008, 'overall_f1': 0.1984517945109078, 'overall_accuracy': 0.6678796960291127}
			------------EPOCH 13---------------
Loss:  tensor(236.8208, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(62.5672, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(262.6429, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(167.5646, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(155.1082, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(485.4500, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(119.3111, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(295.2083, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(208.9784, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(195.9773, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(121.1113, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(188.3314, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.0870, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(246.4608, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(148.9976, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(131.8736, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(80.5444, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(211.6342, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(162.1749, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(69.4743, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(129.0289, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.0612, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.6759, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.1358, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(181.0000, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(183.1279, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(61.0752, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(106.6787, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(218.3826, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(128.8599, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(134.0415, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(448.0594, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(217.4806, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.2308, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.7563, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(268.6028, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(213.6197, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(335.6758, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.1540, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.9151, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(170.4167, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(138.1465, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(362.1987, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.4016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(150.7862, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(121.9675, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(169.4377, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(294.8231, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(162.2503, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(114.0597, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(110.5646, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(139.9661, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(95.1683, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(107.0278, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(82.7233, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.6731, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.5840, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3020833333333333, 'recall': 0.16666666666666666, 'f1': 0.2148148148148148, 'number': 348}, 'P': {'precision': 0.1713747645951036, 'recall': 0.310580204778157, 'f1': 0.220873786407767, 'number': 293}, 'overall_precision': 0.20608575380359612, 'overall_recall': 0.23244929797191888, 'overall_f1': 0.21847507331378296, 'overall_accuracy': 0.6557850797388419}
			------------EPOCH 14---------------
Loss:  tensor(200.5679, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.8838, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(217.4680, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(113.4738, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(136.3445, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(344.8778, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(98.5444, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(213.3027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(222.6410, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(199.8669, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(103.2823, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(201.9125, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.1380, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(201.8330, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(123.1698, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(130.9359, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(64.7897, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(166.9742, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(158.1552, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.4611, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(92.0156, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.7320, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.0251, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.4877, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(127.9717, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(164.3807, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.9831, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(66.3770, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(148.9355, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(85.5792, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(83.0312, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(297.6547, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(134.2984, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.7810, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.9689, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(196.8946, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(167.3191, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(268.9835, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.4435, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.1908, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(122.0665, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(111.7257, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(280.7322, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.0784, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(150.0766, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(94.8475, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(121.2597, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(308.0500, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(176.8865, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(135.2060, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(109.0702, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(142.5456, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(97.0700, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(78.9517, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(75.4518, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.3949, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.2428, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.2560240963855422, 'recall': 0.2442528735632184, 'f1': 0.25, 'number': 348}, 'P': {'precision': 0.20162932790224034, 'recall': 0.3378839590443686, 'f1': 0.25255102040816324, 'number': 293}, 'overall_precision': 0.2235722964763062, 'overall_recall': 0.2870514820592824, 'overall_f1': 0.2513661202185792, 'overall_accuracy': 0.6627421599058119}
			------------EPOCH 15---------------
Loss:  tensor(144.7855, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.1563, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(199.8519, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(98.7821, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(80.9304, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(293.6287, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(66.6117, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(156.4807, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(95.9126, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(103.1778, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(66.5455, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(108.3518, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.7702, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(208.7550, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(118.7639, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(79.6138, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(71.4791, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(141.3469, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(116.5973, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.8158, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(120.4384, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(70.6020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(68.6998, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(71.1469, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(205.8146, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(349.7225, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(81.0579, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(140.9191, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(225.2291, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(122.2209, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(87.2131, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(429.8275, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(135.5743, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.6811, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.2364, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(159.2740, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(128.9681, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(213.5119, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.9771, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.9419, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.2807, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(64.5531, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(207.5672, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.2773, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(108.4792, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(82.7834, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(86.8644, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(189.9212, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(176.2296, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(141.1043, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(129.7580, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(138.3144, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(115.3561, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(122.8767, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(125.8918, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(69.9055, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.2470, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.21875, 'recall': 0.22126436781609196, 'f1': 0.22000000000000003, 'number': 348}, 'P': {'precision': 0.25675675675675674, 'recall': 0.3890784982935154, 'f1': 0.3093622795115332, 'number': 293}, 'overall_precision': 0.2399497487437186, 'overall_recall': 0.29797191887675506, 'overall_f1': 0.26583159359777314, 'overall_accuracy': 0.6378037033072889}
			------------EPOCH 16---------------
Loss:  tensor(245.6762, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(91.4180, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(388.2992, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(141.2494, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(145.8268, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(684.7750, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(277.7144, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(514.3663, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(147.1449, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(194.2436, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(84.5692, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(152.0304, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.3518, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(163.8268, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(89.8790, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(74.0893, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.2094, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(92.8809, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(90.5179, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.2824, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(76.2348, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.1554, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.9979, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.2452, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(223.0628, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(275.6229, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.4065, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(85.5178, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(356.7613, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(211.3756, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(183.7755, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(712.6957, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(244.3074, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(92.5598, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(98.0491, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(381.4461, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(216.6689, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(321.5610, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(75.3411, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.9320, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(107.5316, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(85.6290, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(286.3898, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.8501, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(88.8981, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(71.9734, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(86.5896, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(201.4614, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(110.6103, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(79.1508, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(59.9715, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(153.9124, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(66.7617, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(68.1935, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(89.3776, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.8940, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.8001, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.247557003257329, 'recall': 0.21839080459770116, 'f1': 0.23206106870229007, 'number': 348}, 'P': {'precision': 0.2603036876355748, 'recall': 0.40955631399317405, 'f1': 0.31830238726790455, 'number': 293}, 'overall_precision': 0.2552083333333333, 'overall_recall': 0.30577223088923555, 'overall_f1': 0.27821149751596874, 'overall_accuracy': 0.6411752113882051}
			------------EPOCH 17---------------
Loss:  tensor(194.4095, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.1169, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(201.0057, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(106.2305, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(126.8546, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(402.8392, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(184.9271, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(263.4160, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(257.8878, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(245.1387, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(183.8504, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(172.6949, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(132.0377, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(392.6870, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(296.9676, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(225.5101, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(164.7814, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(300.8323, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(285.7915, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(284.1165, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(202.7832, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(110.0032, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(149.3214, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(112.9679, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(155.2931, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(282.1920, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.6753, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(85.6584, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(196.6211, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(77.7728, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(78.7462, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(319.3876, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(111.9640, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.1736, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.7021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(117.6858, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(123.1214, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(212.3162, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.6115, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.6472, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(104.8119, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(111.3933, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(235.3837, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.7646, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(127.6080, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(169.0323, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(176.1323, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(287.8175, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(216.9870, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(231.4142, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(137.8522, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(175.0495, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(96.1028, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(152.9644, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(141.8462, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.3634, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.6050, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.2788844621513944, 'recall': 0.20114942528735633, 'f1': 0.23372287145242068, 'number': 348}, 'P': {'precision': 0.14260249554367202, 'recall': 0.27303754266211605, 'f1': 0.18735362997658078, 'number': 293}, 'overall_precision': 0.18472906403940886, 'overall_recall': 0.23400936037441497, 'overall_f1': 0.20646937370956642, 'overall_accuracy': 0.6707695600984694}
			------------EPOCH 18---------------
Loss:  tensor(193.7703, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.6331, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(174.0165, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(83.5145, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(71.9179, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(332.0908, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(83.5697, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(137.5090, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(101.4845, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(118.4465, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(101.6040, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(117.1755, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.0122, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(212.7312, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(111.3217, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(59.7907, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.5147, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(96.4939, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(71.8691, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.0025, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(79.4709, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.9254, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.4631, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.1953, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(86.4190, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(128.5818, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.1808, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.9509, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(182.8403, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(65.9328, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(80.7646, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(294.5698, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(127.1291, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.5155, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.5578, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(129.2373, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(114.6855, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(193.7282, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.5768, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.5337, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(59.1164, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(68.7529, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(251.9124, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.6641, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(95.7009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(89.1605, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(138.4554, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(237.7247, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(99.2667, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(71.6986, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.1042, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(103.0780, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.3685, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.0583, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.8801, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.0819, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.0678, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.233983286908078, 'recall': 0.2413793103448276, 'f1': 0.2376237623762376, 'number': 348}, 'P': {'precision': 0.1724137931034483, 'recall': 0.27303754266211605, 'f1': 0.21136063408190225, 'number': 293}, 'overall_precision': 0.19927095990279464, 'overall_recall': 0.25585023400936036, 'overall_f1': 0.2240437158469945, 'overall_accuracy': 0.6627956759070962}
			------------EPOCH 19---------------
Loss:  tensor(129.0836, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.2809, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(94.1118, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.0369, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.5526, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(192.8849, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.2801, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(89.0075, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.1055, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(75.0921, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(67.1696, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(100.7030, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.3773, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(105.5908, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(68.3076, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.0107, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.2861, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(69.6373, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.3005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.6988, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.3470, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.9988, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.6322, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.9712, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.0243, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(92.9521, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.0408, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.2417, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(101.6833, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.0651, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.3107, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(161.2038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(65.2435, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.6021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.2452, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(69.3013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(72.3808, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(146.4875, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.1519, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.6620, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.9741, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.8345, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(118.9422, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.7901, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.0715, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.0012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.1021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(168.5638, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(74.2434, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.2521, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.5304, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(71.8762, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.7689, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.0390, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.2825, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.6416, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.8095, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.2771929824561403, 'recall': 0.22701149425287356, 'f1': 0.2496050552922591, 'number': 348}, 'P': {'precision': 0.22342733188720174, 'recall': 0.3515358361774744, 'f1': 0.27320954907161804, 'number': 293}, 'overall_precision': 0.24396782841823056, 'overall_recall': 0.2839313572542902, 'overall_f1': 0.2624369142033165, 'overall_accuracy': 0.6658996039815905}
			------------EPOCH 20---------------
Loss:  tensor(92.0657, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.5722, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(81.3078, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.7887, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.7714, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(193.7420, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.4756, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(83.6410, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.0784, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.0496, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.3647, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(68.9015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.6421, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(80.1416, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.6517, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.9239, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.6189, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.0250, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.9968, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.2956, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.7349, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.9160, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.9415, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.8778, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.4374, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(62.3167, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.4471, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.2933, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(83.9112, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.0465, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.7453, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(196.8767, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.9481, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.1769, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.3026, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.3728, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.9579, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(140.8760, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.4426, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.1518, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.3697, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.9807, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(104.3655, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.2984, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.1497, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.3876, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.4557, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(150.2634, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(64.6996, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.6450, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.5583, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.2762, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.3616, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.1569, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.8731, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.7991, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.6354, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3046594982078853, 'recall': 0.2442528735632184, 'f1': 0.2711323763955342, 'number': 348}, 'P': {'precision': 0.18003913894324852, 'recall': 0.31399317406143346, 'f1': 0.2288557213930348, 'number': 293}, 'overall_precision': 0.2240506329113924, 'overall_recall': 0.27613104524180965, 'overall_f1': 0.24737945492662475, 'overall_accuracy': 0.6684148560419565}
			------------EPOCH 21---------------
Loss:  tensor(58.6564, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.6729, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(66.3462, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.0782, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.0096, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(187.2899, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.8938, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.5061, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.4738, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.2650, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.8069, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(66.6002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.9533, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.5000, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.1777, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.3132, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.5836, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.2188, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.5912, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.5494, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.7969, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.5622, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.5879, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.9647, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.2807, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.1288, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.1173, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.4356, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(65.2056, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.1061, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.6909, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(175.4854, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(76.7290, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.4583, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.9416, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.0261, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.1291, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(115.0992, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.3523, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.2675, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.9981, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.7364, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(103.2660, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.0388, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.9910, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.9022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.3374, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(73.5696, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(64.6816, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.9466, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.2388, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(75.7268, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.7517, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.7239, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.5421, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.1458, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.5013, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.2443820224719101, 'recall': 0.25, 'f1': 0.2471590909090909, 'number': 348}, 'P': {'precision': 0.21641791044776118, 'recall': 0.29692832764505117, 'f1': 0.25035971223021586, 'number': 293}, 'overall_precision': 0.22955145118733508, 'overall_recall': 0.2714508580343214, 'overall_f1': 0.2487491065046462, 'overall_accuracy': 0.6632773199186557}
			------------EPOCH 22---------------
Loss:  tensor(88.8550, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.4384, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.5472, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.2954, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.2677, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(174.6717, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.4253, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(101.6907, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.0506, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.4913, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.4259, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(69.1112, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.7693, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(98.4706, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.2516, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.0648, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.9953, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.7302, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.9449, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.0182, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.5453, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.6790, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.2933, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.4361, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.0311, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.4478, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.9104, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.0779, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(77.7941, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.8435, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.4619, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(112.0408, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.1664, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.0247, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.8776, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.2865, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.7884, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(93.5301, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.9728, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.3033, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.5616, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.9947, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(72.4504, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.1637, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.2840, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.5173, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(64.6685, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(118.0187, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(83.8329, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.0426, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.8442, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(100.8748, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.7857, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(83.6101, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.8015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.6338, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.2025, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.31226765799256506, 'recall': 0.2413793103448276, 'f1': 0.27228525121555913, 'number': 348}, 'P': {'precision': 0.19215686274509805, 'recall': 0.33447098976109213, 'f1': 0.24408468244084683, 'number': 293}, 'overall_precision': 0.2336328626444159, 'overall_recall': 0.2839313572542902, 'overall_f1': 0.2563380281690141, 'overall_accuracy': 0.6689500160548004}
			------------EPOCH 23---------------
Loss:  tensor(166.0927, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.1559, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(79.3945, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.9613, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.1748, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(284.3394, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.4912, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(97.5898, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.4784, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.0112, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.6859, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.3862, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.7876, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(92.3749, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.9380, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.0327, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.4338, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.9271, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.0697, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.9473, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.7939, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.8073, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.2180, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.5044, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.4841, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.8268, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.1909, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.0018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.9493, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.9059, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.0670, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(108.9639, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.5223, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.4431, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.3336, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.7676, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.5389, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(101.9498, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.7895, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.0228, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.6465, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.2866, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(95.0297, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.5147, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.6479, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.2915, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.9635, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(77.1701, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(88.0324, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.3947, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.9260, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(95.1986, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.1025, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(81.3376, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.4670, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.3533, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.8256, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.28771929824561404, 'recall': 0.23563218390804597, 'f1': 0.2590837282780411, 'number': 348}, 'P': {'precision': 0.19308943089430894, 'recall': 0.3242320819112628, 'f1': 0.24203821656050956, 'number': 293}, 'overall_precision': 0.2277992277992278, 'overall_recall': 0.27613104524180965, 'overall_f1': 0.24964739069111425, 'overall_accuracy': 0.6693781440650755}
			------------EPOCH 24---------------
Loss:  tensor(63.8497, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.6280, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.7350, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.8760, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.2985, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(159.7749, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.0569, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.1487, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.0455, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.7327, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.7057, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.5279, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.1654, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(69.2735, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.5028, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.5799, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.1919, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.3857, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.2712, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.6536, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.5534, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.1453, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.4736, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.3166, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.4597, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.2975, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.5062, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.3736, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(76.5334, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.3518, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.2889, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(114.6676, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(66.4553, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.0477, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.9962, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.7756, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.8104, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(87.1122, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.1649, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.4974, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.3066, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.8295, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.8858, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.1576, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.5781, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.6447, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.0985, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.5867, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.5664, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.2475, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.5101, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.5512, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.3629, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.1254, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.8799, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.5276, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.6244, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.2838283828382838, 'recall': 0.2471264367816092, 'f1': 0.2642089093701997, 'number': 348}, 'P': {'precision': 0.1887966804979253, 'recall': 0.310580204778157, 'f1': 0.23483870967741935, 'number': 293}, 'overall_precision': 0.22547770700636943, 'overall_recall': 0.27613104524180965, 'overall_f1': 0.24824684431977556, 'overall_accuracy': 0.6684683720432409}
			------------EPOCH 25---------------
Loss:  tensor(44.0636, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.8095, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(62.9767, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.3856, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.1021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(132.4798, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.0475, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.3210, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.5189, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.2848, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.5119, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.5010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.7514, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.7305, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.4916, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.8364, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.4364, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.1826, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.7152, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.8117, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.8078, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.0300, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.0395, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.7494, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.1968, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.8250, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.2861, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.2658, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(110.9512, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.1847, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.8494, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(70.2955, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.7015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.2435, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.7179, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.4443, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.8479, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(79.2551, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.9945, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.7617, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.7488, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.6026, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.2065, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.8171, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.0931, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.3597, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.9899, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.2241, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.6609, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.1500, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.0136, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.7180, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.9267, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.0620, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.2775, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.4669, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.1076, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3074074074074074, 'recall': 0.23850574712643677, 'f1': 0.2686084142394822, 'number': 348}, 'P': {'precision': 0.21443736730360935, 'recall': 0.3447098976109215, 'f1': 0.26439790575916233, 'number': 293}, 'overall_precision': 0.2483130904183536, 'overall_recall': 0.2870514820592824, 'overall_f1': 0.2662807525325615, 'overall_accuracy': 0.6693246280637911}
			------------EPOCH 26---------------
Loss:  tensor(38.3600, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.3453, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.4050, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.4406, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.4523, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(119.3722, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.8201, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.6457, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.0325, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.1840, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.3206, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.8783, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5253, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.3510, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.5896, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.8160, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.9904, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.5393, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.3107, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.2113, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.2545, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.1451, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.1541, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.1359, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.6066, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.2294, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.4677, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.9733, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.2094, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.0299, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.2869, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.0011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.4985, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.4018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.9896, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.0450, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.7977, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.3250, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.3148, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.1522, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.8623, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.5067, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.9599, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.1366, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.7899, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.7266, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.4252, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.2771, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.6031, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.8494, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.5706, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.8917, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.2998, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.8296, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.3039, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.1145, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.8669, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.28524590163934427, 'recall': 0.25, 'f1': 0.2664624808575804, 'number': 348}, 'P': {'precision': 0.19375, 'recall': 0.3174061433447099, 'f1': 0.24062095730918498, 'number': 293}, 'overall_precision': 0.22929936305732485, 'overall_recall': 0.28081123244929795, 'overall_f1': 0.2524544179523142, 'overall_accuracy': 0.6673445360162689}
			------------EPOCH 27---------------
Loss:  tensor(45.0523, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.6778, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.8333, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.6019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.2471, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(80.2511, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.1890, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.7270, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.7449, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.4101, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.5918, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.0903, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2088, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.9204, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.5857, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.5381, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.9860, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.1009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.5657, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.4043, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.6325, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.0982, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.8401, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.8138, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.4280, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.4462, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.8420, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.1843, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.9122, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.0578, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.0366, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.3159, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.6373, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.9781, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6661, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.9612, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.3122, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.4915, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.8432, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.7511, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.6236, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.6086, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.9990, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.7251, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.5729, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.7103, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.3314, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.4841, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.2936, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.0469, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.9195, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.2601, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.8045, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.1424, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.5420, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.0577, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5451, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.29553264604810997, 'recall': 0.2471264367816092, 'f1': 0.2691705790297339, 'number': 348}, 'P': {'precision': 0.2053941908713693, 'recall': 0.3378839590443686, 'f1': 0.25548387096774194, 'number': 293}, 'overall_precision': 0.23932729624838292, 'overall_recall': 0.28861154446177845, 'overall_f1': 0.26166902404526166, 'overall_accuracy': 0.67028791608691}
			------------EPOCH 28---------------
Loss:  tensor(28.3028, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.3106, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.1030, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.5144, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.3135, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.0762, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.1243, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.5287, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.3089, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.2895, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.8296, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.3035, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0761, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.9019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.5886, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.4884, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.7013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.7537, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.1343, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.8643, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.1611, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.7052, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.3124, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5519, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.8866, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.9834, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2947, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.7468, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.5630, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.3713, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.5916, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.7440, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.3467, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5442, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3884, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.6198, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.2832, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.4259, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5095, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.4678, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.9260, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.0687, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.6718, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4431, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.7133, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.9774, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.5129, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.2549, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.9708, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.1927, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.1211, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.8258, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.7400, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.1732, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.7468, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.3375, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3053, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.2814569536423841, 'recall': 0.2442528735632184, 'f1': 0.26153846153846155, 'number': 348}, 'P': {'precision': 0.20166320166320167, 'recall': 0.3310580204778157, 'f1': 0.25064599483204136, 'number': 293}, 'overall_precision': 0.23243933588761176, 'overall_recall': 0.2839313572542902, 'overall_f1': 0.2556179775280899, 'overall_accuracy': 0.6685218880445253}
			------------EPOCH 29---------------
Loss:  tensor(20.1344, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.0751, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.3430, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.7657, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.0157, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.1401, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.1144, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.6552, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.2259, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.4157, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.2583, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.6464, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9134, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.4955, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.0462, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.5051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2431, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.6841, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.3177, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.4046, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.7403, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3756, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.8780, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3739, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.5487, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.1542, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.8645, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.3612, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.8805, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.4222, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.9260, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.8175, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.1562, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2837, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1738, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.2468, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.9212, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.1521, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2457, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2143, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.0858, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.6571, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.1992, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2493, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.1243, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.5329, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.2826, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.4717, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.0867, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.8752, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.8890, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.0487, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.2680, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.3866, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.4971, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.9866, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1935, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.2833333333333333, 'recall': 0.2442528735632184, 'f1': 0.2623456790123457, 'number': 348}, 'P': {'precision': 0.2037037037037037, 'recall': 0.3378839590443686, 'f1': 0.25417201540436457, 'number': 293}, 'overall_precision': 0.2340966921119593, 'overall_recall': 0.2870514820592824, 'overall_f1': 0.2578836720392431, 'overall_accuracy': 0.67028791608691}
			------------EPOCH 30---------------
Loss:  tensor(17.3951, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.8314, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.0005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.1809, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5807, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.3516, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5759, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.1208, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.3379, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.2983, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.7960, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.2061, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8460, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.5338, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5270, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.6869, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.0086, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.4438, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.3069, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.1090, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.6678, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2082, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6212, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2187, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.5032, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.7179, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6208, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.1205, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.5356, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.4954, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.1727, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.8558, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.4291, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0962, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0594, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.2091, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.8676, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.6342, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.0415, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.0287, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.7357, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.3867, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.3215, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1264, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.8553, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.4022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.3312, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.2814, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.2958, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.3322, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.4727, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.9176, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.6127, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.0343, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.5398, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5413, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0192, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.2709677419354839, 'recall': 0.2413793103448276, 'f1': 0.2553191489361702, 'number': 348}, 'P': {'precision': 0.19583333333333333, 'recall': 0.32081911262798635, 'f1': 0.24320827943078913, 'number': 293}, 'overall_precision': 0.22531645569620254, 'overall_recall': 0.27769110764430577, 'overall_f1': 0.24877707896575818, 'overall_accuracy': 0.6691640800599379}


		-------------RUN 3-----------
			------------EPOCH 1---------------
Loss:  tensor(2835.8086, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2118.7822, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3460.0793, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2357.0327, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2356.3716, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3522.7017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2349.3784, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1790.5704, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(889.1595, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(959.1334, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1269.3567, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2053.4463, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1274.0527, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2015.2733, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3900.6406, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2178.7502, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1327.5085, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(874.3705, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1661.5688, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1597.3839, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1212.7771, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2237.8889, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1596.6670, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1447.0474, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1688.0886, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2224.2849, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1606.4197, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2939.2314, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2956.7681, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2067.0942, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1642.8901, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1694.6720, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1713.0392, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1618.2654, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1989.6589, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2211.6970, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1759.5159, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2070.9595, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1567.3771, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1668.0737, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1370.9751, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2080.6379, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2199.1404, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1666.2932, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2428.3936, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2434.6660, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1803.7421, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2076.6213, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1052.0536, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1699.4763, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1646.4496, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1386.3120, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2412.0601, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1944.3257, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1099.1625, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1277.4343, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.007194244604316547, 'recall': 0.0037735849056603774, 'f1': 0.004950495049504951, 'number': 265}, 'P': {'precision': 0.0125, 'recall': 0.03557312252964427, 'f1': 0.018499486125385406, 'number': 253}, 'overall_precision': 0.011641443538998836, 'overall_recall': 0.019305019305019305, 'overall_f1': 0.014524328249818447, 'overall_accuracy': 0.5339086827220246}
			------------EPOCH 2---------------
Loss:  tensor(1705.4576, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1228.2471, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2135.1643, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1516.0103, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1669.7996, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2560.8105, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1722.7113, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1412.1256, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(699.4525, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(719.5877, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1000.4646, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1697.4381, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1083.2117, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1651.2496, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3577.9709, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1886.1156, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1110.9415, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(732.3714, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1371.0776, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1225.6921, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1072.1252, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2001.6414, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1372.1404, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1233.8987, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1436.5950, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1783.5042, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1281.1086, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2688.0481, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2483.2288, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1578.2469, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1368.4858, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1500.2040, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1393.9604, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1418.6638, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1740.2576, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2051.7344, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1618.2549, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1806.4031, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1329.6724, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1428.1171, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1239.1267, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1870.2310, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2014.8619, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1460.2858, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2101.9980, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2183.6021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1551.7402, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1888.5020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(907.0016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1486.3064, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1463.9126, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1117.1760, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2020.2079, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1759.9408, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(984.8026, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1045.0642, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.05019305019305019, 'recall': 0.04905660377358491, 'f1': 0.04961832061068702, 'number': 265}, 'P': {'precision': 0.045627376425855515, 'recall': 0.04743083003952569, 'f1': 0.046511627906976744, 'number': 253}, 'overall_precision': 0.04789272030651341, 'overall_recall': 0.04826254826254826, 'overall_f1': 0.04807692307692308, 'overall_accuracy': 0.541430987229576}
			------------EPOCH 3---------------
Loss:  tensor(1485.9728, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(956.1376, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1821.1167, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1383.0470, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1494.2452, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2240.2344, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1497.3467, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1223.7390, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(598.9330, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(635.9188, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(835.2082, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1487.4844, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(939.6482, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1407.9377, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3335.2251, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1696.3865, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(897.3107, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(584.6683, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1134.1801, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(888.5293, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(890.1052, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1719.8762, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1175.9618, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1030.0967, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1138.5685, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1418.9149, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(972.5601, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2078.5591, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2065.8916, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1138.0597, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1113.2507, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1285.5825, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1123.6383, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1244.6371, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1509.5044, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1759.1743, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1392.1107, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1514.0500, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(986.2658, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1170.0991, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1095.5369, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1583.3048, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1711.6392, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1128.3960, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1646.3668, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1893.9950, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1266.1266, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1624.1982, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(791.5915, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1240.0984, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1227.5363, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(836.1009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1506.5718, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1407.7285, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(743.1256, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(858.2383, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.0648854961832061, 'recall': 0.06415094339622641, 'f1': 0.06451612903225806, 'number': 265}, 'P': {'precision': 0.055900621118012424, 'recall': 0.07114624505928854, 'f1': 0.06260869565217392, 'number': 253}, 'overall_precision': 0.059931506849315065, 'overall_recall': 0.06756756756756757, 'overall_f1': 0.06352087114337568, 'overall_accuracy': 0.5897136859292087}
			------------EPOCH 4---------------
Loss:  tensor(1186.6907, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(647.2937, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1405.6785, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1025.7990, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1230.7368, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1760.8439, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1182.3174, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1001.0707, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(423.3655, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(525.9297, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(683.7596, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1149.1404, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(724.2250, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1095.2146, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2680.5327, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1324.9109, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(681.9062, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(440.5149, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(837.3813, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(547.1780, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(709.0116, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1352.5549, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(898.0536, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(809.6555, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(848.3875, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1115.1355, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(669.0476, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1491.0660, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1638.3640, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(877.6495, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(850.8196, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(992.9132, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(787.6025, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1022.1317, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1145.9539, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1410.3040, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1440.7073, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1245.7662, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(707.1273, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(883.8750, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(856.0774, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1245.0623, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1388.0596, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(793.5039, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1129.5278, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1418.3660, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(990.6606, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1297.6140, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(667.9921, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(932.5060, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(878.4865, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(641.5817, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1261.5308, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1116.7349, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(562.7469, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(660.3464, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.15916955017301038, 'recall': 0.17358490566037735, 'f1': 0.16606498194945848, 'number': 265}, 'P': {'precision': 0.15269461077844312, 'recall': 0.2015810276679842, 'f1': 0.17376490630323682, 'number': 253}, 'overall_precision': 0.15569823434991975, 'overall_recall': 0.18725868725868725, 'overall_f1': 0.1700262927256792, 'overall_accuracy': 0.6517581200069975}
			------------EPOCH 5---------------
Loss:  tensor(1000.1348, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(355.3563, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1029.1899, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(818.5630, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(999.1678, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1422.5776, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1027.2167, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(874.5601, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(275.3528, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(441.7095, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(504.9281, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(984.6295, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(478.9894, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(902.1794, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2147.3804, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(855.8351, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(453.3087, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(354.6653, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(624.5437, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(341.0485, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(599.7933, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1224.5273, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(855.1143, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(669.6449, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(631.8129, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(809.1694, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(415.9290, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1209.5099, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1352.8472, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(704.1836, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(618.0865, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(710.7296, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(605.7208, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(981.4099, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1075.6436, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1244.7124, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(805.7160, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(914.5833, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(585.8800, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(730.7916, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(633.8097, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(928.1183, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1056.1503, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(463.5341, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(959.6467, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1312.2515, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(912.4264, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1302.4562, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(558.4846, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(850.4834, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(746.7818, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(590.0932, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(991.6960, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(791.4731, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(422.3876, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(548.1632, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.18049792531120332, 'recall': 0.3283018867924528, 'f1': 0.2329317269076305, 'number': 265}, 'P': {'precision': 0.07738095238095238, 'recall': 0.05138339920948617, 'f1': 0.061757719714964375, 'number': 253}, 'overall_precision': 0.15384615384615385, 'overall_recall': 0.19305019305019305, 'overall_f1': 0.17123287671232879, 'overall_accuracy': 0.573502828153245}
			------------EPOCH 6---------------
Loss:  tensor(1171.7822, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(317.9902, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1153.3508, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1172.5013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1370.7476, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2203.7588, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1095.5179, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1082.2583, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(233.8137, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(363.1710, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(364.4457, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(760.2147, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(331.8598, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(786.5446, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1800.4962, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(737.2076, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(378.3542, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(332.7297, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(604.0186, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(278.0105, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(568.9069, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1262.4955, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(864.8666, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(741.9707, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(620.2583, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(722.6947, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(347.2906, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1315.2651, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1286.9425, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(596.5660, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(559.3399, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(731.4399, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(605.8945, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(736.6686, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(851.5145, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1123.9930, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(736.2721, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(647.1544, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(343.0498, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(512.6746, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(746.3064, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(914.4187, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(949.6650, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(498.6906, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1139.0017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1139.4034, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(755.2157, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1015.2388, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(501.1605, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(680.1124, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(583.1123, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(445.5897, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(767.8443, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(599.8147, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(305.1914, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(470.8765, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.20502092050209206, 'recall': 0.18490566037735848, 'f1': 0.19444444444444442, 'number': 265}, 'P': {'precision': 0.11468812877263582, 'recall': 0.22529644268774704, 'f1': 0.152, 'number': 253}, 'overall_precision': 0.14402173913043478, 'overall_recall': 0.20463320463320464, 'overall_f1': 0.16905901116427433, 'overall_accuracy': 0.6708262872470698}
			------------EPOCH 7---------------
Loss:  tensor(734.5254, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(243.0619, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(733.2309, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(549.5215, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(789.6816, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1034.8365, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(704.4276, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(610.1385, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(169.9913, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(350.9033, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(340.8796, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(623.6515, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(242.6271, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(605.0495, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1242.6289, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(515.2487, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(281.8298, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(246.8368, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(426.5829, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(232.7262, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(381.7397, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(749.1724, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(560.0988, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(508.8231, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(329.6862, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(513.3783, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(241.8684, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(785.2099, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(943.1291, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(466.9504, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(364.5636, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(409.5653, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(344.7950, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(523.1816, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(597.7359, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(745.6350, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(563.3844, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(417.8785, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(199.1360, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(341.1575, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(431.4157, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(563.2090, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(640.6705, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(227.0160, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(511.5491, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(604.5248, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(454.9210, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(885.3083, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(369.0181, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(459.8634, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(393.2359, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(301.7435, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(517.0682, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(384.6200, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(190.2439, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(341.4161, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.20552147239263804, 'recall': 0.2528301886792453, 'f1': 0.22673434856175975, 'number': 265}, 'P': {'precision': 0.07754010695187166, 'recall': 0.11462450592885376, 'f1': 0.09250398724082935, 'number': 253}, 'overall_precision': 0.13714285714285715, 'overall_recall': 0.18532818532818532, 'overall_f1': 0.15763546798029557, 'overall_accuracy': 0.6806810892763426}
			------------EPOCH 8---------------
Loss:  tensor(543.4586, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(128.5260, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(545.3931, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(451.4003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(484.8439, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(768.4137, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(579.0109, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(503.8856, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(127.8882, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(243.8005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(156.2480, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(509.7231, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(203.0780, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(490.5017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1042.7335, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(394.7440, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(160.1574, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(239.9682, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(337.6490, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(176.9319, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(283.6036, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(535.8032, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(442.6482, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(437.1881, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(212.7898, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(315.6273, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(141.4836, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(555.4742, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(665.9045, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(388.4505, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(271.7662, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(255.9193, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(250.9233, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(360.6117, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(477.1614, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(560.8112, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(552.8162, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(383.5066, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(182.3622, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(322.3650, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(282.6230, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(389.2026, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(531.1931, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(155.9620, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(356.3604, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(321.5977, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(334.2272, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(588.6861, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(285.0893, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(387.6175, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(311.3650, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(257.5424, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(498.4552, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(341.7347, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(188.0357, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(263.5371, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.19631901840490798, 'recall': 0.24150943396226415, 'f1': 0.21658206429780033, 'number': 265}, 'P': {'precision': 0.1721311475409836, 'recall': 0.33201581027667987, 'f1': 0.22672064777327933, 'number': 253}, 'overall_precision': 0.18181818181818182, 'overall_recall': 0.2857142857142857, 'overall_f1': 0.2222222222222222, 'overall_accuracy': 0.6315820164441075}
			------------EPOCH 9---------------
Loss:  tensor(515.5298, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(114.4524, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(596.0168, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(484.3388, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(454.1997, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(709.4766, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(453.7262, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(488.2853, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(108.6609, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(212.2022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(125.0243, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(484.4417, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(158.0350, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(353.8161, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(739.4521, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(306.3551, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(157.5901, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(227.4528, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(321.8576, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(259.1976, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(235.6852, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(496.4373, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(412.3810, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(372.2036, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(196.1718, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(314.0099, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(144.6631, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(711.6688, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(546.6430, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(318.3135, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(233.8636, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(222.4641, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(231.2098, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(257.5139, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(375.2233, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(417.9164, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(478.1859, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(278.2910, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(129.0809, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(213.8181, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(250.7487, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(446.9178, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(574.7911, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(101.4098, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(248.1880, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(198.6221, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(270.0943, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(523.9567, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(284.1914, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(402.8973, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(365.5258, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(257.0379, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(414.5298, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(236.9843, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(110.0141, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(181.9543, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.18882978723404256, 'recall': 0.2679245283018868, 'f1': 0.22152886115444617, 'number': 265}, 'P': {'precision': 0.1928374655647383, 'recall': 0.2766798418972332, 'f1': 0.2272727272727273, 'number': 253}, 'overall_precision': 0.19079837618403248, 'overall_recall': 0.2722007722007722, 'overall_f1': 0.2243436754176611, 'overall_accuracy': 0.6588139250102046}
			------------EPOCH 10---------------
Loss:  tensor(324.0390, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(66.2405, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(303.8872, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(306.0364, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(375.0412, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(547.5906, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(326.0788, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(301.1724, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(99.7555, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(188.9267, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(95.6865, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(389.8217, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(119.0468, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(310.7302, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(786.6432, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(232.8367, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(62.4767, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(214.4854, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(228.8250, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(118.8819, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(165.2206, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(328.4376, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(292.6113, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(292.4728, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(113.4700, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(203.0226, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(100.2375, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(408.4263, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(406.6529, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(220.8867, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(160.3395, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(193.8720, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(195.6678, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(211.0555, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(322.6508, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(594.3106, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(286.9294, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(209.4498, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(85.7285, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(193.0043, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(141.0290, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(393.3640, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(444.2401, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(110.6203, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(215.7831, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(188.6978, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(246.8550, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(381.5173, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(227.6910, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(246.4204, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(224.5501, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(179.9801, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(329.3121, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(187.4780, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(104.4118, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(159.7166, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.23444976076555024, 'recall': 0.18490566037735848, 'f1': 0.20675105485232068, 'number': 265}, 'P': {'precision': 0.21001926782273603, 'recall': 0.4308300395256917, 'f1': 0.2823834196891192, 'number': 253}, 'overall_precision': 0.21703296703296704, 'overall_recall': 0.305019305019305, 'overall_f1': 0.25361155698234356, 'overall_accuracy': 0.657531051373258}
			------------EPOCH 11---------------
Loss:  tensor(317.6247, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.5204, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(355.1517, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(298.5534, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(305.9892, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(467.6494, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(268.9128, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(272.7666, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(78.0506, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(159.5724, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(84.4338, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(343.7414, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(117.7750, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(274.6363, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(775.7520, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(235.5348, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.1210, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(170.4846, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(289.3086, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(118.4057, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(140.4646, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(269.5196, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(311.4961, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(212.7816, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(131.5492, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(188.6490, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(74.3240, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(386.4473, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(368.2781, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(300.7190, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(210.0789, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(170.4345, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(147.9346, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(212.9091, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(319.8057, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(488.3520, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(234.3931, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(242.8793, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(92.3525, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(188.2014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(122.7808, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(262.6377, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(398.5629, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(104.4930, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(258.7506, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(225.9929, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(320.7661, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(475.3077, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(201.5234, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(229.6236, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(209.8593, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(194.8675, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(262.9824, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(136.4826, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(69.3688, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(115.5707, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.30128205128205127, 'recall': 0.17735849056603772, 'f1': 0.2232779097387173, 'number': 265}, 'P': {'precision': 0.19230769230769232, 'recall': 0.43478260869565216, 'f1': 0.26666666666666666, 'number': 253}, 'overall_precision': 0.21565934065934067, 'overall_recall': 0.3030888030888031, 'overall_f1': 0.2520064205457464, 'overall_accuracy': 0.6516414951309114}
			------------EPOCH 12---------------
Loss:  tensor(335.4183, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(68.9844, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(348.6882, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(244.0099, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(376.4037, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(596.9463, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(264.5980, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(372.9328, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(76.4846, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(179.9758, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(100.8137, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(434.1489, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(100.9252, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(281.8767, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(898.9965, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(286.4233, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.7241, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(149.7295, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(191.5253, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(127.2724, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(137.9516, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(298.4958, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(310.0493, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(179.3369, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(123.7627, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(201.9613, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(76.3140, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(462.0170, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(449.5260, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(215.2207, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(214.0484, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(156.9700, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(159.1251, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(281.6047, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(322.9033, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(475.2850, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(223.0352, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(218.7333, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(115.9090, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(162.7776, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(151.6773, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(236.5442, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(466.4168, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(101.5670, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(377.6786, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(283.8548, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(317.7530, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(498.5042, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(327.7723, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(204.4492, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(218.7037, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(266.3382, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(371.2467, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(212.5711, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(138.1034, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(187.5306, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.32323232323232326, 'recall': 0.12075471698113208, 'f1': 0.1758241758241758, 'number': 265}, 'P': {'precision': 0.12849872773536897, 'recall': 0.39920948616600793, 'f1': 0.1944177093358999, 'number': 253}, 'overall_precision': 0.15028248587570622, 'overall_recall': 0.25675675675675674, 'overall_f1': 0.1895937277263008, 'overall_accuracy': 0.615662720858359}
			------------EPOCH 13---------------
Loss:  tensor(241.6391, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(77.1262, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(305.5988, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(254.1383, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(340.6045, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(590.5292, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(299.1660, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(326.4838, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(108.5289, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(332.1725, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(326.9684, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(525.2039, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(221.6740, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(426.0673, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(904.3701, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(369.3297, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(100.0226, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(148.7484, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(291.3138, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(210.5644, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(148.9347, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(345.1725, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(302.7184, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(196.4646, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(127.6420, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(241.5359, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(137.1579, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(664.8080, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(780.4152, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(435.1876, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(305.5353, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(271.5371, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(252.4916, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(432.3433, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(513.1501, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(513.3409, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(348.3636, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(319.1899, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(138.9757, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(283.1511, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(254.6011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(283.3851, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(485.6953, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(163.1047, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(367.9173, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(284.4036, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(395.9885, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(503.4827, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(269.0846, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(186.6253, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(176.8331, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(265.2610, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(664.2094, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(542.2326, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(167.0586, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(514.9448, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.21014492753623187, 'recall': 0.10943396226415095, 'f1': 0.14392059553349876, 'number': 265}, 'P': {'precision': 0.1382716049382716, 'recall': 0.22134387351778656, 'f1': 0.1702127659574468, 'number': 253}, 'overall_precision': 0.15653775322283608, 'overall_recall': 0.1640926640926641, 'overall_f1': 0.16022620169651272, 'overall_accuracy': 0.6683771648492624}
			------------EPOCH 14---------------
Loss:  tensor(356.2252, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(90.6582, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(446.4318, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(390.4769, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(303.7795, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(388.2711, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(193.6080, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(228.5942, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.2881, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(115.7842, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(76.9911, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(307.8115, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.4501, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(196.4616, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(641.8293, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(220.2862, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(557.5045, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(355.3000, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(482.4593, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(325.6254, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(419.0272, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1509.1221, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(741.7676, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(426.1925, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(329.0913, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(297.8678, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(102.2868, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1045.4519, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(666.5939, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(292.3009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(230.8381, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(175.4041, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(278.7393, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(411.3753, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(314.1989, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(467.0778, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(590.3007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(453.5266, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(341.7212, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(409.0225, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(788.3527, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(615.6646, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(747.5013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(511.7929, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(928.3372, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(733.9374, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(425.4213, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(639.3387, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(271.2747, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(243.2949, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(197.7068, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(264.2546, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(354.0223, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(265.1536, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(159.2274, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(208.0478, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.367816091954023, 'recall': 0.12075471698113208, 'f1': 0.18181818181818182, 'number': 265}, 'P': {'precision': 0.08759124087591241, 'recall': 0.09486166007905138, 'f1': 0.09108159392789374, 'number': 253}, 'overall_precision': 0.15512465373961218, 'overall_recall': 0.10810810810810811, 'overall_f1': 0.1274175199089875, 'overall_accuracy': 0.6429529418624993}
			------------EPOCH 15---------------
Loss:  tensor(593.5858, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(212.0250, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(480.5040, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(392.5914, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(610.2067, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(881.9868, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(581.2729, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(540.0173, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(287.4603, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(223.3489, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(317.0673, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(888.8595, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(100.8108, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(414.7331, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1373.4348, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(333.4649, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(177.3366, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(122.7575, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(254.2401, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(120.5128, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(228.3487, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(532.0364, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(413.9370, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(298.3782, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(282.6064, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(251.2860, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(123.8575, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(790.5306, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(726.8043, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(448.2161, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(294.4915, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(288.8346, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(242.2189, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(601.7493, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(588.1304, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(481.6092, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(681.7052, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(425.2510, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(153.3419, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(304.3801, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(473.1007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(242.0380, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(524.3736, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(601.9275, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(426.0381, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(639.0964, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(244.8938, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(571.9875, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(191.1796, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(178.3507, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(139.5033, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(385.0034, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(712.7065, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(495.4584, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(299.5292, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(226.6871, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.15409309791332262, 'recall': 0.3622641509433962, 'f1': 0.2162162162162162, 'number': 265}, 'P': {'precision': 0.06278026905829596, 'recall': 0.05533596837944664, 'f1': 0.058823529411764705, 'number': 253}, 'overall_precision': 0.13002364066193853, 'overall_recall': 0.21235521235521235, 'overall_f1': 0.16129032258064516, 'overall_accuracy': 0.6270336462767508}
			------------EPOCH 16---------------
Loss:  tensor(657.8885, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(155.4843, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(673.0546, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(979.2836, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(672.6851, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1025.1611, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(378.1312, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(463.4000, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(198.4525, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(238.0146, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(162.3540, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(577.3480, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(142.1933, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(366.9859, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(733.9596, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(308.0269, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(130.8684, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(213.6756, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(248.0861, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(151.0589, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(162.4625, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(355.9831, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(308.5210, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(191.0754, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(170.1731, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(191.8565, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(65.2527, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(498.5598, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(481.2263, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(189.3145, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(132.1206, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(158.1323, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(177.8143, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(237.4117, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(309.5747, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(615.1799, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(459.7692, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(331.0249, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(196.3309, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(190.2330, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(160.7600, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(722.3021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(814.5668, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(163.9874, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(408.7108, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(338.4977, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(362.3210, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1075.4014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(290.8788, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(245.3962, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(258.3851, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(286.5259, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(357.5509, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(172.2583, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(142.1252, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(109.0250, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.1968503937007874, 'recall': 0.2830188679245283, 'f1': 0.23219814241486067, 'number': 265}, 'P': {'precision': 0.17251461988304093, 'recall': 0.233201581027668, 'f1': 0.19831932773109245, 'number': 253}, 'overall_precision': 0.18533886583679116, 'overall_recall': 0.25868725868725867, 'overall_f1': 0.21595487510072522, 'overall_accuracy': 0.6501836841798355}
			------------EPOCH 17---------------
Loss:  tensor(224.9665, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.9294, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(284.9335, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(263.6857, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(245.0665, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(391.0176, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(246.4916, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(226.3787, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(72.2349, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(106.9581, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(68.9497, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(367.6668, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(123.1608, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(334.7004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(699.3909, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(250.5059, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(105.6742, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(96.8552, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(251.0730, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(227.8434, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(179.1632, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(285.6483, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(322.5523, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(179.1000, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(131.5699, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(299.0475, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(83.6344, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(540.6798, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(405.8890, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(469.6107, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(204.8152, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(129.6563, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(108.6327, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(267.7529, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(272.7717, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(313.2941, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(144.7169, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(150.7177, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.7726, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(115.5318, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.7311, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(155.2381, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(256.3706, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.7404, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(149.0502, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(84.8325, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(171.1532, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(272.9371, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(117.3846, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(71.8942, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(83.0698, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(115.1293, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(184.3920, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(75.2152, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.9011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(65.8401, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.22040816326530613, 'recall': 0.2037735849056604, 'f1': 0.21176470588235294, 'number': 265}, 'P': {'precision': 0.14623655913978495, 'recall': 0.26877470355731226, 'f1': 0.1894150417827298, 'number': 253}, 'overall_precision': 0.17183098591549295, 'overall_recall': 0.23552123552123552, 'overall_f1': 0.1986970684039088, 'overall_accuracy': 0.6917021400664762}
			------------EPOCH 18---------------
Loss:  tensor(166.1746, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.0094, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(134.9784, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(111.0773, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(200.2887, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(329.9349, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(164.3850, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(127.5806, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.0080, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(73.3387, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.3837, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(219.5005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.7273, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(124.2281, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(440.4899, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(142.4756, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.4429, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.8537, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(85.9511, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.9727, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(80.3818, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(101.6475, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(160.8320, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(88.8499, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.2282, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(84.2205, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.8693, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(289.8018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(197.6727, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(123.6736, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(74.2532, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.6963, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(69.5741, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(124.9164, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(141.1254, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(232.8259, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(142.2510, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(94.3794, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.4344, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(69.2191, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.2660, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(94.9638, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(198.6653, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.9162, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(87.6262, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.7374, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(106.7666, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(219.6218, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(101.0099, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.8529, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.6187, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(81.2633, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(120.0437, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.8510, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.1017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.3827, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.23674911660777384, 'recall': 0.2528301886792453, 'f1': 0.24452554744525548, 'number': 265}, 'P': {'precision': 0.18446601941747573, 'recall': 0.30039525691699603, 'f1': 0.2285714285714286, 'number': 253}, 'overall_precision': 0.20575539568345325, 'overall_recall': 0.27606177606177607, 'overall_f1': 0.23577906018136852, 'overall_accuracy': 0.6901277042393142}
			------------EPOCH 19---------------
Loss:  tensor(113.4948, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.6093, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(98.9802, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(77.7643, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(135.4246, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(218.3379, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(86.6495, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(65.9062, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.6666, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.4433, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.3976, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(147.4136, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.6771, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(84.5043, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(278.5027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(71.0527, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.4536, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.1379, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(65.6120, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.8318, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.7787, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(70.9297, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(110.8386, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.9622, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.3699, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(64.8520, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.2371, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(199.2032, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(145.5494, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(97.5862, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.7750, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.3416, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.7715, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(84.5174, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(83.1974, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(167.3613, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(85.0396, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(66.4358, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.9367, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.8684, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.9943, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(93.7822, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(170.3162, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.1724, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.4567, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.0926, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(84.5481, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(140.9774, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(72.4337, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.9985, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.1136, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.4115, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(87.6845, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.0816, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.0393, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.0935, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.21518987341772153, 'recall': 0.25660377358490566, 'f1': 0.23407917383821, 'number': 265}, 'P': {'precision': 0.18863049095607234, 'recall': 0.2885375494071146, 'f1': 0.22812500000000002, 'number': 253}, 'overall_precision': 0.20056899004267426, 'overall_recall': 0.2722007722007722, 'overall_f1': 0.23095823095823095, 'overall_accuracy': 0.6866289579567322}
			------------EPOCH 20---------------
Loss:  tensor(92.5093, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.6671, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(85.5389, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.6590, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(96.6426, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(163.5978, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(62.1381, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.1419, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.6432, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.4949, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.5685, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(114.4672, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.1965, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.9922, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(191.8988, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.9901, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.4994, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.8645, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.0549, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.3539, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.7579, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.5541, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(86.0952, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.5472, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.0183, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.9855, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.6609, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(145.0142, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(126.9659, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(80.5345, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.9946, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.7603, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.2487, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(66.5462, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(74.1922, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(145.3794, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(66.7462, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.2306, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.8013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.1803, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.0896, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(61.3483, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(131.6416, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.9648, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.5838, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.5236, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(66.6594, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(110.9327, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.5324, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.5441, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.9284, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.3408, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(64.1902, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.3028, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.2185, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.5269, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.21753246753246752, 'recall': 0.2528301886792453, 'f1': 0.2338568935427574, 'number': 265}, 'P': {'precision': 0.18575063613231552, 'recall': 0.2885375494071146, 'f1': 0.22600619195046437, 'number': 253}, 'overall_precision': 0.19971469329529243, 'overall_recall': 0.2702702702702703, 'overall_f1': 0.22969647251845773, 'overall_accuracy': 0.6837716484926235}
			------------EPOCH 21---------------
Loss:  tensor(78.0954, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.2897, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(71.9827, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.8275, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(71.2275, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(138.1456, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.2191, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.5543, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.5761, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.3394, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.5995, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(94.6388, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.9921, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.4522, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(144.9836, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.3910, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.5462, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.8157, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.0892, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.0318, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.2910, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.7957, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(68.2039, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.5670, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.5895, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.1913, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.1961, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(116.3398, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(109.1155, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(66.5647, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.0844, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.2448, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.9378, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.6246, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.0700, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(121.8744, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.5106, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.9927, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.9955, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.9658, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.2109, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.1431, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(109.1763, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.7235, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.5900, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.7271, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.0105, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(89.0390, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.8190, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.5631, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.0404, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.0091, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.1882, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.0481, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.6830, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.9666, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.21474358974358973, 'recall': 0.2528301886792453, 'f1': 0.23223570190641246, 'number': 265}, 'P': {'precision': 0.1889168765743073, 'recall': 0.2964426877470356, 'f1': 0.23076923076923078, 'number': 253}, 'overall_precision': 0.2002820874471086, 'overall_recall': 0.27413127413127414, 'overall_f1': 0.23145884270578646, 'overall_accuracy': 0.6844713977491399}
			------------EPOCH 22---------------
Loss:  tensor(69.3562, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.8177, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(65.1415, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.5133, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.8850, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(117.3088, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.4719, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.8709, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.1903, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.2581, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.6162, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(83.4787, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.3815, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.6767, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(122.0728, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.7545, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.0410, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.4201, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.6603, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.6891, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.6820, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.4629, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.3047, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.4486, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.0227, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.2982, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.2263, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(100.0051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(96.7472, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.6866, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.3540, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.4701, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.3238, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.9123, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.3265, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(111.4431, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.9871, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.6248, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.6319, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.3869, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.7211, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.3694, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(91.8951, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.8170, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.2312, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.8380, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.1475, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(75.8436, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.4358, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.0765, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.2025, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.4860, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.3100, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.3102, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.4443, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.3090, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.20967741935483872, 'recall': 0.24528301886792453, 'f1': 0.22608695652173913, 'number': 265}, 'P': {'precision': 0.17929292929292928, 'recall': 0.28063241106719367, 'f1': 0.21879815100154082, 'number': 253}, 'overall_precision': 0.19263456090651557, 'overall_recall': 0.2625482625482625, 'overall_f1': 0.22222222222222218, 'overall_accuracy': 0.6828969619219779}
			------------EPOCH 23---------------
Loss:  tensor(57.7386, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.6589, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.9631, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.2562, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.4804, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(125.9348, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(65.9728, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.9904, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.7722, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.8274, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.2976, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(68.4517, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.1685, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.4623, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(103.6043, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.7778, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.2206, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.2588, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.9785, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.1107, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.2962, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.6041, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.1932, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.7497, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.6240, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.9152, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.7874, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(92.0480, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(87.4420, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.6326, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.2030, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.0139, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.0865, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.3238, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.3695, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(108.6166, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.9034, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.3080, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.0241, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.2402, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.6809, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.0170, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(84.2781, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.6895, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.1732, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.6220, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.0067, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(71.4997, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.4388, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.1067, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.5571, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.7836, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.7728, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.6276, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.5432, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.8064, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.2168284789644013, 'recall': 0.2528301886792453, 'f1': 0.23344947735191635, 'number': 265}, 'P': {'precision': 0.1761786600496278, 'recall': 0.28063241106719367, 'f1': 0.21646341463414634, 'number': 253}, 'overall_precision': 0.19382022471910113, 'overall_recall': 0.26640926640926643, 'overall_f1': 0.22439024390243903, 'overall_accuracy': 0.6837716484926235}
			------------EPOCH 24---------------
Loss:  tensor(54.7965, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.6898, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.6074, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.1113, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.3779, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(128.9532, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(93.8812, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.7345, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.3473, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.4103, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.1515, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.6290, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.6365, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.4307, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(107.7464, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.9395, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.3412, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.1690, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.9802, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.0811, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.4168, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.1676, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.2088, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.3470, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.6094, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.5740, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.7596, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(95.8140, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(80.4597, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.7130, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.7122, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.3796, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.8801, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.0709, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.1208, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(101.5923, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.0215, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.0248, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.0503, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.2940, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.7902, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.2464, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(85.7942, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.6503, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.9305, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.5323, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.9841, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(65.9685, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.9632, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.2472, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.2692, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.9287, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.9256, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.3771, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.2386, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.1918, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.20915032679738563, 'recall': 0.24150943396226415, 'f1': 0.22416812609457093, 'number': 265}, 'P': {'precision': 0.1683673469387755, 'recall': 0.2608695652173913, 'f1': 0.20465116279069767, 'number': 253}, 'overall_precision': 0.18624641833810887, 'overall_recall': 0.25096525096525096, 'overall_f1': 0.21381578947368418, 'overall_accuracy': 0.6816724007230742}
			------------EPOCH 25---------------
Loss:  tensor(49.3495, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.9340, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.8712, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.8056, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.7568, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(84.4145, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.3205, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.7443, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.2887, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.1667, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.1195, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.3945, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.1628, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.1417, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(123.2729, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.3943, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.9139, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.5722, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.6482, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.8483, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.9557, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.1656, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.4073, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.7422, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.2466, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.8023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.0424, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(79.5493, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(71.3707, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.6064, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.3706, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.6367, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.0504, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.0963, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.3343, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(88.1736, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.3006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.8040, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.8785, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.7483, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.9280, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.5795, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(65.5551, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.0413, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.8476, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.0096, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.5864, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.5182, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.3533, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.4088, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.7022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.4987, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.3537, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.0717, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.0124, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.3041, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.21103896103896103, 'recall': 0.24528301886792453, 'f1': 0.2268760907504363, 'number': 265}, 'P': {'precision': 0.1728395061728395, 'recall': 0.2766798418972332, 'f1': 0.2127659574468085, 'number': 253}, 'overall_precision': 0.18934081346423562, 'overall_recall': 0.2606177606177606, 'overall_f1': 0.21933387489845652, 'overall_accuracy': 0.6806227768382996}
			------------EPOCH 26---------------
Loss:  tensor(47.7767, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.2846, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.7597, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.6075, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.4183, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(66.2735, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.6958, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.9571, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.7318, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.6321, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.5919, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.0963, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.5669, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.7467, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(80.8509, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.6417, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.4108, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.7238, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.6756, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.7405, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.4973, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.2602, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.0241, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.1610, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.1110, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.7546, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.2187, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(69.6515, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(66.0064, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.2250, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.1138, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.5519, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.9488, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.7500, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.8597, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(79.7821, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.3540, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.3205, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.0336, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.3490, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.7585, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.6218, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(61.0865, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.1916, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.0381, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.1332, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.4076, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.8222, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.8550, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.6377, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.7107, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.4869, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.8156, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.6029, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.0571, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.6575, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.22333333333333333, 'recall': 0.2528301886792453, 'f1': 0.23716814159292035, 'number': 265}, 'P': {'precision': 0.17042606516290726, 'recall': 0.26877470355731226, 'f1': 0.20858895705521474, 'number': 253}, 'overall_precision': 0.19313304721030042, 'overall_recall': 0.2606177606177606, 'overall_f1': 0.2218570254724733, 'overall_accuracy': 0.6817307131611172}
			------------EPOCH 27---------------
Loss:  tensor(46.8861, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.7668, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.2530, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.3333, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.2349, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.1761, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.6853, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.4491, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.3969, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.4012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.7358, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.9851, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.1446, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.6352, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.1014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.1113, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.4424, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.7042, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.2641, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.6270, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.8785, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.3697, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.6927, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.7901, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.5679, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.4076, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.6360, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.3209, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.0491, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.9527, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.1022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.9013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.5775, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.1344, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.4141, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(74.7551, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.8741, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.6925, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.2571, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.0325, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.2543, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.2794, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.4408, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.6125, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.1960, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.3724, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.9548, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.0237, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.6397, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.1620, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.7349, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.7246, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.6020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.1782, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.4071, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.4613, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.21476510067114093, 'recall': 0.24150943396226415, 'f1': 0.22735346358792188, 'number': 265}, 'P': {'precision': 0.1722488038277512, 'recall': 0.2845849802371542, 'f1': 0.21460506706408347, 'number': 253}, 'overall_precision': 0.18994413407821228, 'overall_recall': 0.2625482625482625, 'overall_f1': 0.22042139384116693, 'overall_accuracy': 0.6733920345209633}
			------------EPOCH 28---------------
Loss:  tensor(43.1541, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.3108, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.5938, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.1027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.6268, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.2048, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.9805, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.6948, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.7463, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.5214, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.3303, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.1861, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.3970, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.2503, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.1322, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.7138, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.0283, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.8432, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.2094, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.0156, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.2377, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.8131, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.9617, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.3776, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.9274, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.1300, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.1755, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.4259, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.0526, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.6366, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.7470, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.2736, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.5280, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.6328, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.5212, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(80.1165, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.7277, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.9914, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.7907, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.5459, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.6205, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.4958, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.9577, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.1891, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.5904, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.3165, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.0691, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.8905, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.9769, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.7514, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.2670, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.6745, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.1970, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.0666, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.1429, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.6808, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.20521172638436483, 'recall': 0.23773584905660378, 'f1': 0.22027972027972026, 'number': 265}, 'P': {'precision': 0.17530864197530865, 'recall': 0.28063241106719367, 'f1': 0.21580547112462006, 'number': 253}, 'overall_precision': 0.18820224719101122, 'overall_recall': 0.25868725868725867, 'overall_f1': 0.2178861788617886, 'overall_accuracy': 0.6746749081579101}
			------------EPOCH 29---------------
Loss:  tensor(33.2840, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.0794, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.9871, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.4608, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.0655, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.7535, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.8120, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.5781, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2919, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.5096, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.0852, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.6389, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.9280, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.2359, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.7675, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.0822, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.7503, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.2017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.1227, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.6875, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.6566, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.9628, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.8498, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.3493, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.4854, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.6153, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.8869, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.2865, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.6471, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.3364, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.8150, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.8386, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.7841, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.1317, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.4967, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(86.1330, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.5309, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.0406, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.3360, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.4447, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.3475, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.0199, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.6607, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.7021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.3873, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.9423, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.2967, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.4984, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.6692, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5763, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.8259, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.6411, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.9696, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.3960, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.1668, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.2108843537414966, 'recall': 0.2339622641509434, 'f1': 0.22182468694096602, 'number': 265}, 'P': {'precision': 0.176056338028169, 'recall': 0.2964426877470356, 'f1': 0.22091310751104565, 'number': 253}, 'overall_precision': 0.19027777777777777, 'overall_recall': 0.2644787644787645, 'overall_f1': 0.22132471728594508, 'overall_accuracy': 0.6765409061752872}
			------------EPOCH 30---------------
Loss:  tensor(25.5995, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6894, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.6813, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.0371, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.5245, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.9317, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.6870, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.0820, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.1159, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.8215, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.5533, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.9444, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.6609, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.4608, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.7207, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.1256, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4504, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.8150, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.5778, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2969, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.6136, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.5490, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.4488, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.5823, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.3216, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.4166, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6063, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.7948, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.1572, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.0458, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.6382, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.4771, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.2799, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.3168, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.9708, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(62.9057, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.0643, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.0688, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.1282, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.5528, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.9560, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.7235, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(59.0297, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.4839, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.7726, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.1998, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.1857, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.9571, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.7530, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2416, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5809, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.2135, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.0061, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.8182, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.4645, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.7884, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.21639344262295082, 'recall': 0.2490566037735849, 'f1': 0.23157894736842102, 'number': 265}, 'P': {'precision': 0.19597989949748743, 'recall': 0.308300395256917, 'f1': 0.23963133640552997, 'number': 253}, 'overall_precision': 0.20483641536273114, 'overall_recall': 0.277992277992278, 'overall_f1': 0.23587223587223588, 'overall_accuracy': 0.6732170972068342}


		-------------RUN 4-----------
			------------EPOCH 1---------------
Loss:  tensor(4166.4224, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3322.8052, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2917.8770, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2822.9373, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2146.0254, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3591.0791, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1931.7787, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1950.4326, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2299.3147, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2408.9868, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3479.0137, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2990.8560, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2554.0803, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2383.0510, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2177.1562, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2675.8843, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1120.2119, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2428.4937, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2169.1780, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1652.5909, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2380.9995, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1588.6831, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1995.4050, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1861.1909, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1680.9115, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2449.3442, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2380.7534, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1434.6166, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1495.1903, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1544.9531, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2276.0596, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1115.7478, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1858.4210, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2248.0391, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1043.0322, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(853.2084, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1537.3818, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2299.4487, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2191.1465, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2031.7139, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1653.4808, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1599.7411, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1126.4431, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1229.7917, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1074.5225, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1139.3176, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1698.9099, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1795.5125, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1938.7629, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2325.8591, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1720.8434, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1497.0319, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1440.3329, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2058.2803, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1280.9688, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1479.0497, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1395.6075, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1649.1531, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1420.8121, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.02973977695167286, 'recall': 0.037209302325581395, 'f1': 0.03305785123966942, 'number': 215}, 'P': {'precision': 0.05235602094240838, 'recall': 0.07604562737642585, 'f1': 0.06201550387596899, 'number': 263}, 'overall_precision': 0.043010752688172046, 'overall_recall': 0.058577405857740586, 'overall_f1': 0.049601417183348095, 'overall_accuracy': 0.5959101006141383}
			------------EPOCH 2---------------
Loss:  tensor(2922.7432, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2269.1274, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1893.7759, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1758.1619, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1405.9128, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2541.4170, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1296.0563, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1299.7573, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1668.2101, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1787.3198, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2735.4761, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2374.1907, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1991.6345, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1824.5134, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1541.0978, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1828.8123, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(821.6334, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2125.9829, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1696.7544, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1297.5176, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2169.4585, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1299.3735, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1549.6316, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1391.4998, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1327.5505, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2065.9258, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2050.0088, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1145.4517, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1174.5239, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1180.4348, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1746.1218, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(889.3948, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1506.3722, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2029.0039, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(813.1284, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(617.3656, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1230.7640, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1812.9999, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1842.3596, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1693.5581, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1359.8711, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1331.7406, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(889.6637, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(975.5546, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(881.9795, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(909.7909, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1343.2876, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1369.6794, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1585.4214, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1933.4056, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1360.9915, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1199.7439, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1189.9982, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1615.0106, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1096.4700, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1156.1416, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1034.4678, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1320.3420, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1182.6023, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.10619469026548672, 'recall': 0.22325581395348837, 'f1': 0.14392803598200898, 'number': 215}, 'P': {'precision': 0.19343065693430658, 'recall': 0.20152091254752852, 'f1': 0.1973929236499069, 'number': 263}, 'overall_precision': 0.13911845730027547, 'overall_recall': 0.2112970711297071, 'overall_f1': 0.16777408637873753, 'overall_accuracy': 0.6320397229844505}
			------------EPOCH 3---------------
Loss:  tensor(2519.5820, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1996.7678, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1583.1196, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1539.3516, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1203.7661, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2170.2378, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1101.4135, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1073.1576, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1389.9204, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1506.9336, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2235.1792, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1909.1687, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1601.3250, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1551.0415, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1274.6530, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1472.7042, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(621.0785, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1700.7443, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1361.8853, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1077.7615, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1832.6965, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1096.8906, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1293.1929, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1207.6814, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1127.9421, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1638.0188, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1781.3337, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(943.3676, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1027.0112, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(991.0695, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1495.4509, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(770.5465, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1186.8231, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1621.1885, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(646.8641, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(466.1909, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(970.9736, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1500.5222, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1569.2078, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1351.2896, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1021.6340, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1057.9427, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(660.1235, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(738.4623, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(736.6487, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(801.2578, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1075.4561, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1134.2322, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1322.8470, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1578.9148, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(943.8585, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(913.8510, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(984.5833, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1145.6365, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(875.0040, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(926.2367, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(810.4092, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1059.5308, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1010.1338, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.09898477157360407, 'recall': 0.1813953488372093, 'f1': 0.12807881773399016, 'number': 215}, 'P': {'precision': 0.17478510028653296, 'recall': 0.23193916349809887, 'f1': 0.19934640522875818, 'number': 263}, 'overall_precision': 0.13458950201884254, 'overall_recall': 0.20920502092050208, 'overall_f1': 0.1638001638001638, 'overall_accuracy': 0.662485299882399}
			------------EPOCH 4---------------
Loss:  tensor(2213.4209, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1693.7758, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1304.0500, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1222.9474, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1000.7327, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1891.0139, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(945.4560, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(876.7814, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1148.3292, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1200.8724, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1695.2737, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1424.7432, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1237.8047, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1369.9395, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(996.3350, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1148.9448, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(462.8676, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1120.0001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1092.6652, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(880.4534, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1433.3435, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(823.9827, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1064.9493, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1016.5436, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1010.9709, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1219.9465, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1444.3132, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(742.8011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(876.2122, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(790.4712, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1291.6731, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(654.0806, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(966.0956, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1317.0134, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(505.9841, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(376.7782, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(799.1451, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1295.5398, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1302.0352, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(993.7944, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(741.4059, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(881.4312, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(466.5185, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(526.4318, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(614.5322, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(703.5158, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(969.5443, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(993.3748, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1144.4271, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1371.9923, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(715.3207, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(716.7150, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(872.1297, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(890.1498, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(708.9408, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(702.7317, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(510.4825, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(822.0511, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(898.9608, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.11061946902654868, 'recall': 0.23255813953488372, 'f1': 0.14992503748125935, 'number': 215}, 'P': {'precision': 0.12461059190031153, 'recall': 0.1520912547528517, 'f1': 0.136986301369863, 'number': 263}, 'overall_precision': 0.11642949547218628, 'overall_recall': 0.18828451882845187, 'overall_f1': 0.14388489208633093, 'overall_accuracy': 0.660329282634261}
			------------EPOCH 5---------------
Loss:  tensor(1583.2471, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1300.7855, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(837.4424, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1106.2211, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(835.0543, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1654.9722, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(859.2353, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(731.7219, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1017.5694, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(988.3167, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1325.0524, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(982.7367, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1060.4821, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1227.8334, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(789.2101, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(901.1049, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(378.7632, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(997.9718, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(898.9762, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(835.7204, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1325.0231, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(742.2361, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(973.8209, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(955.4879, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1026.9451, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1638.0546, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1668.0938, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(631.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(733.1653, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(580.5679, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1079.9506, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(427.9027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(941.1577, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1187.4470, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(510.3922, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(372.8602, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(791.6042, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1412.2454, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1430.9836, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1285.3538, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(800.9650, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(870.5728, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(468.9402, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(519.8900, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(551.5273, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(533.7706, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(798.6946, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(796.7827, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(958.4718, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1192.8473, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(517.2218, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(489.8120, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(770.0745, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(846.0554, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(571.7146, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(674.5700, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(555.0012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(774.9775, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(802.8324, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.21726190476190477, 'recall': 0.3395348837209302, 'f1': 0.26497277676951, 'number': 215}, 'P': {'precision': 0.3096590909090909, 'recall': 0.4144486692015209, 'f1': 0.3544715447154471, 'number': 263}, 'overall_precision': 0.26453488372093026, 'overall_recall': 0.3807531380753138, 'overall_f1': 0.3121783876500858, 'overall_accuracy': 0.6664053312426499}
			------------EPOCH 6---------------
Loss:  tensor(1689.8389, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1261.9811, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(928.8400, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(939.0220, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(773.4836, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1550.3904, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(973.2384, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(822.8945, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(986.7803, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1038.2664, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1517.4778, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1499.6985, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(893.7197, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1054.2480, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(699.1338, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(846.0249, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(316.6490, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(800.4833, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(790.6938, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(704.4509, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1114.7794, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(749.2089, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(878.9807, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(957.7996, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(850.2717, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1147.9766, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1434.5962, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(664.2824, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(842.3818, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(793.6573, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1326.4269, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(488.2946, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(689.5364, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(935.8880, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(344.0547, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(300.6725, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(592.8845, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(890.0420, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(896.1149, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(682.8280, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(569.1431, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(713.0538, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(304.8733, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(381.5616, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(703.2015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(664.7834, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(971.4913, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1003.1976, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1117.0404, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1699.9768, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(791.3429, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1069.2346, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(711.0931, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1021.0612, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(443.1744, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(602.5914, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(326.1882, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(601.0249, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(602.3949, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.14715719063545152, 'recall': 0.20465116279069767, 'f1': 0.17120622568093385, 'number': 215}, 'P': {'precision': 0.1570048309178744, 'recall': 0.24714828897338403, 'f1': 0.19202363367799113, 'number': 263}, 'overall_precision': 0.15287517531556802, 'overall_recall': 0.2280334728033473, 'overall_f1': 0.18303946263643994, 'overall_accuracy': 0.7034496275970208}
			------------EPOCH 7---------------
Loss:  tensor(1338.6602, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(990.9908, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(669.8932, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(712.2192, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(635.9077, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1471.2698, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(726.0563, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(572.0520, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1140.3948, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1000.2698, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1266.9723, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(948.0110, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(667.5317, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(864.2673, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(578.5270, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(758.8447, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(276.0855, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(649.2533, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(655.6926, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(696.7965, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(934.4142, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(547.9231, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(820.5315, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(772.4952, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(673.8828, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(950.2675, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1019.0565, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(487.4793, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(555.2674, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(461.5653, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(879.1066, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(293.5520, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(584.5218, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(751.9749, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(265.3290, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(226.6993, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(474.6180, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(623.9149, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(688.6616, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(529.4589, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(407.9143, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(441.0211, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(216.1081, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(217.7348, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(469.0485, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(340.2449, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(620.7484, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(565.0910, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(808.7148, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1135.7151, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(466.1054, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(525.1902, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(496.7770, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(787.4218, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(296.4067, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(545.0867, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(262.4334, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(555.3452, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(458.9346, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.11180124223602485, 'recall': 0.16744186046511628, 'f1': 0.1340782122905028, 'number': 215}, 'P': {'precision': 0.12598425196850394, 'recall': 0.18250950570342206, 'f1': 0.14906832298136646, 'number': 263}, 'overall_precision': 0.11948790896159317, 'overall_recall': 0.17573221757322174, 'overall_f1': 0.14225232853513972, 'overall_accuracy': 0.7103096824774598}
			------------EPOCH 8---------------
Loss:  tensor(989.4180, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(875.6715, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(533.2278, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(735.5619, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(381.2233, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(995.2745, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(508.0325, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(392.3777, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(521.4578, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(461.5696, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(652.1786, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(542.7424, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(406.4047, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(569.4691, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(371.1921, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(618.5745, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(222.6079, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(507.6848, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(489.7244, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(570.1165, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(725.2406, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(472.1316, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(688.1925, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(737.5213, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(707.8519, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1199.8369, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1253.6292, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(476.0744, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(586.8976, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(424.0516, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(826.7290, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(230.5755, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(412.5020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(842.6636, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(216.0888, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(162.6566, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(461.9836, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(421.5963, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(514.2469, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(561.1865, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(389.5501, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(466.9140, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(141.9006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(151.2606, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(360.3860, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(255.4484, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(466.9092, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(416.6968, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(565.2774, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(855.5656, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(342.6242, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(271.4799, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(331.5318, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(550.2129, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(215.7264, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(400.5249, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(201.6545, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(456.5950, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(465.7809, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.11153846153846154, 'recall': 0.13488372093023257, 'f1': 0.12210526315789474, 'number': 215}, 'P': {'precision': 0.03409090909090909, 'recall': 0.045627376425855515, 'f1': 0.03902439024390244, 'number': 263}, 'overall_precision': 0.06699346405228758, 'overall_recall': 0.08577405857740586, 'overall_f1': 0.07522935779816513, 'overall_accuracy': 0.639749117992944}
			------------EPOCH 9---------------
Loss:  tensor(945.5693, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1092.6472, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(635.9212, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(877.8817, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(412.0630, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1136.8390, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(530.2979, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(407.0964, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(520.8494, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(396.8095, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(685.8164, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(491.4094, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(294.9384, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(458.8668, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(351.3259, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(501.5457, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(168.8903, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(423.5477, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(389.3086, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(418.1512, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(606.9695, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(347.0349, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(570.3920, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(611.8029, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(508.2232, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(823.7708, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(744.4348, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(550.3235, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(687.0330, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(414.4382, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(873.1950, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(216.6471, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(370.0580, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1004.6606, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(183.3151, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(194.9863, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(481.4174, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(411.9444, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(716.2649, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(567.7979, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(396.3038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(647.2272, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(135.1078, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(199.5051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(208.0848, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(192.2883, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(480.5757, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(339.7831, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(450.8211, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(680.3502, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(371.8391, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(288.0084, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(341.7852, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(371.7375, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(160.9443, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(217.8929, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(134.1312, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(242.7692, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(297.4182, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.12449799196787148, 'recall': 0.14418604651162792, 'f1': 0.13362068965517243, 'number': 215}, 'P': {'precision': 0.04603580562659847, 'recall': 0.06844106463878327, 'f1': 0.05504587155963303, 'number': 263}, 'overall_precision': 0.0765625, 'overall_recall': 0.10251046025104603, 'overall_f1': 0.08765652951699465, 'overall_accuracy': 0.6922122043643015}
			------------EPOCH 10---------------
Loss:  tensor(688.7281, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(607.2224, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(377.8795, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(496.9869, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(357.4758, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(879.5562, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(378.4835, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(314.6372, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(553.4677, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(418.2236, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(630.3867, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(499.5513, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(287.7872, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(531.4150, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(387.6466, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(445.2662, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(157.1095, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(346.7710, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(366.0782, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(316.2540, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(669.5652, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(253.0581, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(405.3979, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(505.0970, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(417.3628, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(411.1863, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(449.4067, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(243.4447, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(313.1774, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(223.7188, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(561.8823, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(135.6232, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(209.8937, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(439.9561, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(90.5903, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(101.7046, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(343.0452, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(222.2245, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(384.9736, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(398.1984, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(308.1784, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(526.3368, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(108.5018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(124.1481, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(170.7873, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(213.4850, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(408.1202, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(396.4900, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(457.2535, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(642.8691, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(417.8175, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(346.5500, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(589.9731, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(664.1395, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(324.3223, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(269.6732, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(133.2006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(315.1033, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(361.8348, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.22636103151862463, 'recall': 0.3674418604651163, 'f1': 0.2801418439716312, 'number': 215}, 'P': {'precision': 0.24731182795698925, 'recall': 0.34980988593155893, 'f1': 0.28976377952755905, 'number': 263}, 'overall_precision': 0.23717059639389737, 'overall_recall': 0.3577405857740586, 'overall_f1': 0.28523769808173477, 'overall_accuracy': 0.6932575460603685}
			------------EPOCH 11---------------
Loss:  tensor(1030.7555, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(569.0797, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(407.6747, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(397.3552, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(271.7743, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(718.4813, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(335.0592, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(216.7671, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(353.6190, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(241.6730, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(392.6771, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(357.9454, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(191.8837, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(324.8787, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(234.3524, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(325.0377, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(133.3915, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(276.9678, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(292.5098, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(295.8234, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(516.3054, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(276.8881, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(401.5099, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(467.0154, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(415.9706, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(420.1489, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(499.6544, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(243.6390, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(307.5660, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(212.9942, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(660.9618, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(153.6795, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(204.5616, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(344.8151, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(80.2600, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(148.3993, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(344.5211, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(383.7280, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(441.7529, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(281.8318, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(277.5804, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(285.1960, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(115.5124, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(143.8793, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(241.3542, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(134.4095, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(292.9461, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(248.6732, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(314.5186, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(598.6289, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(223.7989, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(151.7066, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(241.3380, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(341.6445, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(77.5859, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(148.8179, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(67.9251, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(127.4709, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(186.6408, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.21676300578034682, 'recall': 0.3488372093023256, 'f1': 0.26737967914438504, 'number': 215}, 'P': {'precision': 0.1388101983002833, 'recall': 0.18631178707224336, 'f1': 0.1590909090909091, 'number': 263}, 'overall_precision': 0.17739628040057226, 'overall_recall': 0.2594142259414226, 'overall_f1': 0.2107051826677995, 'overall_accuracy': 0.6988109238207239}
			------------EPOCH 12---------------
Loss:  tensor(484.4351, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(394.6237, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(293.6503, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(242.1210, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(201.4623, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(571.7655, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(273.2324, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(174.2600, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(280.2504, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(143.0202, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(256.3311, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(253.4813, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(111.0767, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(224.6749, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(167.9459, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(224.9127, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(78.7695, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(220.4889, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(205.5474, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(214.8390, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(402.7192, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(177.0921, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(235.1104, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(359.8905, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(284.0298, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(252.9644, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(358.6957, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(171.9701, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(268.7223, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(235.8031, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(481.8469, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(106.5473, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(135.6720, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(256.2051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(67.7211, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(76.0752, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(244.0975, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(140.9995, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(233.7997, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(184.1388, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(181.9507, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(152.9461, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.4975, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(67.9249, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(98.5235, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(86.5731, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(189.3739, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(159.3014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(220.4937, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(386.8190, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(130.8527, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(98.5691, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(175.6579, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(235.3652, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.2983, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(104.0348, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(93.0124, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(123.0316, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(181.6204, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.1541501976284585, 'recall': 0.3627906976744186, 'f1': 0.21636615811373094, 'number': 215}, 'P': {'precision': 0.08532423208191127, 'recall': 0.09505703422053231, 'f1': 0.08992805755395683, 'number': 263}, 'overall_precision': 0.12891113892365458, 'overall_recall': 0.21548117154811716, 'overall_f1': 0.16131558339859045, 'overall_accuracy': 0.6551679080099307}
			------------EPOCH 13---------------
Loss:  tensor(384.7869, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(356.2477, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(229.7824, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(396.5750, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(199.9456, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(489.0560, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(226.8028, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(175.9484, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(222.8634, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(126.9862, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(239.5307, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(221.5944, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(79.9279, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(199.1675, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(135.4226, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(138.3464, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.5380, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(205.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(229.4439, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(198.3970, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(306.2434, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(154.7924, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(199.7321, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(293.4139, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(223.4861, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(213.9773, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(293.3108, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(132.5849, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(159.6187, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(121.6080, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(336.3444, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(71.3513, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(105.6566, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(269.8409, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.2649, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.3054, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(211.7218, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(143.1923, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(251.5881, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(218.1670, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(154.8372, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(154.1350, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.6228, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(75.2767, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(88.0188, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(72.9857, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(157.6602, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(144.1673, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(161.9292, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(304.6233, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(82.9741, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(64.4611, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(158.3667, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(160.8546, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.4221, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(77.1932, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.4689, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(65.5006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(94.8952, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.18674698795180722, 'recall': 0.4325581395348837, 'f1': 0.2608695652173913, 'number': 215}, 'P': {'precision': 0.09876543209876543, 'recall': 0.09125475285171103, 'f1': 0.09486166007905138, 'number': 263}, 'overall_precision': 0.15789473684210525, 'overall_recall': 0.24476987447698745, 'overall_f1': 0.19196062346185397, 'overall_accuracy': 0.6539918986018555}
			------------EPOCH 14---------------
Loss:  tensor(309.5743, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(292.3440, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(200.9454, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(233.2932, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(158.5777, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(511.8605, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(210.7805, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(152.2997, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(331.0321, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(200.9229, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(359.5081, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(334.4955, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(99.0027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(234.1988, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(167.0018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(113.4407, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.9076, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(157.5710, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(174.6563, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(184.4607, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(217.1603, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(130.4776, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(142.2721, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(209.8820, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(172.5285, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(145.0919, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(245.4412, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(100.1581, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(109.7197, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(115.7025, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(298.8752, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(61.9130, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(172.0483, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(223.6823, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.1524, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(61.2449, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(236.3235, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(444.4512, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(391.5256, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(213.1350, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(150.1381, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(107.6339, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(74.4200, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(76.8452, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(98.9764, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(110.1562, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(385.5247, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(220.1080, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(205.3096, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(377.1094, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(61.2257, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(59.0019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(138.1247, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(151.6283, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.3329, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(82.2170, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.3825, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(71.6311, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(74.3735, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.17233560090702948, 'recall': 0.35348837209302325, 'f1': 0.23170731707317077, 'number': 215}, 'P': {'precision': 0.15613382899628253, 'recall': 0.1596958174904943, 'f1': 0.15789473684210528, 'number': 263}, 'overall_precision': 0.16619718309859155, 'overall_recall': 0.24686192468619247, 'overall_f1': 0.19865319865319864, 'overall_accuracy': 0.6690186854828172}
			------------EPOCH 15---------------
Loss:  tensor(276.4282, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(235.3150, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(213.9079, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(142.9752, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(130.1072, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(394.5598, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(162.0172, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(134.3016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(269.4272, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(213.5775, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(308.9196, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(262.9435, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(587.6681, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(685.6871, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(518.0020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(629.7573, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(74.4018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(191.8415, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(153.7672, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(175.7334, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(249.7388, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(135.0079, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(186.0203, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(313.7135, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(186.8955, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(185.2047, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(237.6615, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(89.8060, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(82.0180, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(110.5478, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(258.9939, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.2766, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(119.2109, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(231.6470, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(113.9178, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(88.8376, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(223.8962, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(285.4035, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(322.8023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(243.7451, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(174.4607, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(167.1109, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(108.9508, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(143.4638, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(208.1123, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(390.1931, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(210.0206, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(174.1925, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(296.1628, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(342.8734, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(104.7946, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(229.6957, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(172.7333, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(292.9715, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(84.5246, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(315.5254, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.1497, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(119.4442, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(91.4394, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.17751479289940827, 'recall': 0.27906976744186046, 'f1': 0.216998191681736, 'number': 215}, 'P': {'precision': 0.2066115702479339, 'recall': 0.28517110266159695, 'f1': 0.23961661341853036, 'number': 263}, 'overall_precision': 0.19258202567760344, 'overall_recall': 0.2824267782426778, 'overall_f1': 0.22900763358778628, 'overall_accuracy': 0.70031360250882}
			------------EPOCH 16---------------
Loss:  tensor(349.7215, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(347.5493, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(234.3247, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(168.0874, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(119.1641, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(467.5974, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(132.6488, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(105.9258, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(270.4065, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(146.0849, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(305.0435, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(207.2919, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(122.0433, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(320.3430, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(265.3000, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(156.9565, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(139.1177, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(365.4776, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(503.4825, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(463.5467, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1060.1279, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(665.4909, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(792.5149, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(615.3411, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(559.6273, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(434.0737, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(471.4969, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(419.3403, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(260.7947, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(188.7960, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(539.3865, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(216.8346, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(164.4087, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(234.5883, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(139.8173, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(64.9779, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(234.2520, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(314.2901, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(366.6625, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(164.6172, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(113.4098, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(130.8717, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.6309, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(68.3555, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(99.9372, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(98.4585, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(155.9559, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(158.2217, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(259.1503, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(380.4788, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(104.7598, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(121.0905, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(291.6811, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(233.3470, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(114.9630, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(477.4289, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(494.2941, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(494.3862, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(531.4515, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.2336448598130841, 'recall': 0.11627906976744186, 'f1': 0.15527950310559005, 'number': 215}, 'P': {'precision': 0.20454545454545456, 'recall': 0.4448669201520912, 'f1': 0.28023952095808385, 'number': 263}, 'overall_precision': 0.20913107511045656, 'overall_recall': 0.29707112970711297, 'overall_f1': 0.24546240276577358, 'overall_accuracy': 0.680125441003528}
			------------EPOCH 17---------------
Loss:  tensor(856.4642, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(834.2021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1040.0923, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(617.4688, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(214.7924, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1093.1167, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(414.3928, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(319.2599, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(271.4955, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(215.2770, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(365.5384, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(219.2033, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(111.4998, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(164.8379, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(121.7303, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(99.6448, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.6132, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(139.3365, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(109.0844, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(146.5082, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(232.8022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(154.3951, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(177.8658, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(247.7322, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(247.7861, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(266.9217, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(354.3970, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(245.7675, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(253.1947, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(410.5798, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(611.5907, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(217.9899, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(983.2008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(885.0394, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(246.3158, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(156.6526, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(580.1462, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(545.0732, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(698.1833, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(612.8519, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(212.7016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(379.9447, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.4850, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(98.8650, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(130.6060, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(118.2354, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(237.1045, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(224.6217, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(188.0408, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(362.4190, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(93.0579, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(131.8103, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(113.2343, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(163.5483, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.5381, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(148.2292, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(69.2335, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(120.6186, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(131.9335, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.1527777777777778, 'recall': 0.15348837209302327, 'f1': 0.15313225058004643, 'number': 215}, 'P': {'precision': 0.16981132075471697, 'recall': 0.2737642585551331, 'f1': 0.2096069868995633, 'number': 263}, 'overall_precision': 0.1640625, 'overall_recall': 0.2196652719665272, 'overall_f1': 0.1878354203935599, 'overall_accuracy': 0.7086110022213511}
			------------EPOCH 18---------------
Loss:  tensor(394.8938, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(321.4023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(265.9506, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(171.2470, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(179.9078, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(505.5522, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(186.9895, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(132.2041, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(389.2527, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(185.3237, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(372.9166, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(301.4128, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(422.2831, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(512.0842, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(491.5806, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(186.3163, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.4763, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(220.3889, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(233.7018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(252.6102, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(353.1409, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(173.4500, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(295.6389, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(445.4709, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(165.5475, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(147.6142, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(247.8772, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(85.6984, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(79.1121, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(89.3344, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(232.6746, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.9536, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.1397, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(123.8997, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.8264, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.4496, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(122.3830, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(78.2216, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(153.9262, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(102.9732, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(87.1529, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(156.8698, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.7783, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.3668, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.0464, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(73.8961, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(135.9583, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(106.0494, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(156.3456, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(375.9565, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(85.4519, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(100.2586, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(163.2521, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(189.3332, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.9318, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(135.1599, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.0253, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(144.7733, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(81.8443, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.14439655172413793, 'recall': 0.3116279069767442, 'f1': 0.19734904270986744, 'number': 215}, 'P': {'precision': 0.0670926517571885, 'recall': 0.07984790874524715, 'f1': 0.07291666666666667, 'number': 263}, 'overall_precision': 0.11325611325611326, 'overall_recall': 0.18410041841004185, 'overall_f1': 0.1402390438247012, 'overall_accuracy': 0.6882921730040507}
			------------EPOCH 19---------------
Loss:  tensor(218.2193, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(197.7252, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(114.5223, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(261.4647, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(80.9537, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(300.5233, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(146.8681, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(89.3738, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(152.2584, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(70.2186, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(184.4940, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(132.8015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.3862, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(145.8731, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(109.4604, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(135.9818, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.1032, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(109.9964, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(75.3490, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(103.8418, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(161.9034, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(75.2357, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(81.2421, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(162.1997, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(120.1987, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(122.1819, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(169.5733, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.1020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.1825, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(65.9808, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(168.4406, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.1598, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.1477, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(104.5784, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.4365, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.8297, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(90.8108, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(101.2805, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(146.1409, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(111.0204, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(68.1359, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(82.9511, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.0955, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.8190, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.8564, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(62.4757, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(107.1126, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(117.5253, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(142.5989, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(246.8235, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.5044, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.8979, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(69.4503, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(66.2946, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.4084, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.6233, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.1072, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.7712, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.8977, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.23986486486486486, 'recall': 0.3302325581395349, 'f1': 0.27788649706457924, 'number': 215}, 'P': {'precision': 0.22538860103626943, 'recall': 0.33079847908745247, 'f1': 0.26810477657935283, 'number': 263}, 'overall_precision': 0.2316715542521994, 'overall_recall': 0.3305439330543933, 'overall_f1': 0.27241379310344827, 'overall_accuracy': 0.7112243564615184}
			------------EPOCH 20---------------
Loss:  tensor(183.0714, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(166.8268, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(110.0309, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(84.0680, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.0644, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(247.6442, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(75.8986, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.3061, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(137.2633, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.1714, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(133.2844, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.4921, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.5733, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(92.4963, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.0738, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.8447, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.4014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(92.0057, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(72.7461, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(82.0233, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(130.3363, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.3172, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(64.4455, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(128.0925, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(105.6946, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(108.1995, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(133.4908, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.6628, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.6614, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(61.5838, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(138.8105, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.0691, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.6383, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.4036, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.1302, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.9274, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.3538, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(68.0937, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(155.4957, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.4500, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.9436, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(94.9685, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.4498, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(59.1313, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.2042, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.9950, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.9901, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.9938, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(84.1801, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(170.0833, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.4114, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.1136, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.7752, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(77.6259, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.4487, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.0479, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.8016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.8470, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.1171, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.20923076923076922, 'recall': 0.31627906976744186, 'f1': 0.2518518518518518, 'number': 215}, 'P': {'precision': 0.14769975786924938, 'recall': 0.23193916349809887, 'f1': 0.1804733727810651, 'number': 263}, 'overall_precision': 0.17479674796747968, 'overall_recall': 0.2698744769874477, 'overall_f1': 0.21217105263157898, 'overall_accuracy': 0.7202404285900954}
			------------EPOCH 21---------------
Loss:  tensor(116.4498, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(114.5229, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(75.1643, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(61.9076, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.6957, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(184.8431, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(68.1324, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.3341, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(107.7651, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.5892, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(100.4006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.6364, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.5136, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(76.0993, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.2771, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.7121, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.7917, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(78.2066, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(80.1255, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(71.2559, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(155.8841, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.3614, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.3041, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(120.1331, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(83.6851, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(83.9089, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(119.5219, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.2807, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.5535, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.2758, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(103.1976, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.2074, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.9559, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.6874, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.1036, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.4357, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.1672, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.5629, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(91.2372, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.5039, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.9472, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.5723, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.5804, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.3611, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.5179, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.9496, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.2245, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.7598, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.8362, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(140.3281, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.2032, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.8765, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.3525, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.8386, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.8766, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.1288, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.3006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.8939, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.9819, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.22333333333333333, 'recall': 0.3116279069767442, 'f1': 0.26019417475728157, 'number': 215}, 'P': {'precision': 0.1272264631043257, 'recall': 0.19011406844106463, 'f1': 0.1524390243902439, 'number': 263}, 'overall_precision': 0.16883116883116883, 'overall_recall': 0.24476987447698745, 'overall_f1': 0.19982920580700256, 'overall_accuracy': 0.7241604599503463}
			------------EPOCH 22---------------
Loss:  tensor(107.3373, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(101.4697, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(69.9977, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.9166, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.7237, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(176.2424, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.6314, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.7597, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(83.7175, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.0030, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(93.7415, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.5695, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.9851, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(70.3327, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.6406, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.4426, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.9566, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(83.4369, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.8609, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.1836, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(116.8029, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.6541, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.9078, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(96.5267, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(69.7893, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(71.7964, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(84.2077, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.4749, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.4621, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.1576, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(78.2614, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.0894, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.2987, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.5397, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.1388, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.9501, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.4296, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.2132, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(67.9372, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.7365, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.3041, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.7903, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.3975, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.3099, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.4712, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.3983, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.5113, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.0638, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.5891, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(121.0980, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.3041, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.8408, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.2432, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.3544, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.3535, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.2586, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.3648, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.4679, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.9892, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.2297734627831715, 'recall': 0.3302325581395349, 'f1': 0.2709923664122137, 'number': 215}, 'P': {'precision': 0.18811881188118812, 'recall': 0.2889733840304182, 'f1': 0.22788605697151423, 'number': 263}, 'overall_precision': 0.2061711079943899, 'overall_recall': 0.3075313807531381, 'overall_f1': 0.2468513853904282, 'overall_accuracy': 0.7146870508297399}
			------------EPOCH 23---------------
Loss:  tensor(109.7834, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(90.9512, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.7184, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.2636, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.1426, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(176.1631, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.7033, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.5978, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(70.8422, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.6734, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(70.2535, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.9718, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.5868, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(68.4295, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.3742, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.5691, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.8197, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(73.2767, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.9462, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.6900, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(79.3525, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.8201, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.1157, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(72.7996, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.5037, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.9760, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(72.8810, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.9518, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.6577, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.9619, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(59.5722, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.5129, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.7632, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.7819, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.9585, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.7923, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.8201, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.3494, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(65.5025, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.0930, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.3001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.9246, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.9940, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.6854, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.3820, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.2358, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.2830, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.1093, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.1024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(100.4179, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.1096, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.0807, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.6390, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(61.6955, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.7917, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.8087, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.2110, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.4234, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.3847, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.2098360655737705, 'recall': 0.29767441860465116, 'f1': 0.24615384615384614, 'number': 215}, 'P': {'precision': 0.17336683417085427, 'recall': 0.2623574144486692, 'f1': 0.20877458396369136, 'number': 263}, 'overall_precision': 0.1891891891891892, 'overall_recall': 0.27824267782426776, 'overall_f1': 0.2252328535139712, 'overall_accuracy': 0.7157323925258069}
			------------EPOCH 24---------------
Loss:  tensor(83.4961, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(77.5601, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.7901, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.6091, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.0404, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(115.8829, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.0079, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.6666, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.0176, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.7865, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.2954, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.6260, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.7333, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.5122, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.6758, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.2114, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.3902, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.3051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.0327, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.1070, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.9224, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.6201, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.8255, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(64.9398, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.6542, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.5755, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.4634, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.4924, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.0555, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.6097, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.8652, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.2784, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.9545, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.6629, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.7150, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.6292, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.6030, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.8090, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(59.0270, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.6885, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.7206, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.8928, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.7533, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.6346, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.5548, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.0563, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.8357, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.6623, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.8180, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(77.2387, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.6671, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.2070, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.0735, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.3373, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.6218, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.4604, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.9894, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.7255, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.4023, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.23432343234323433, 'recall': 0.3302325581395349, 'f1': 0.2741312741312741, 'number': 215}, 'P': {'precision': 0.181360201511335, 'recall': 0.2737642585551331, 'f1': 0.21818181818181817, 'number': 263}, 'overall_precision': 0.2042857142857143, 'overall_recall': 0.29916317991631797, 'overall_f1': 0.24278438030560268, 'overall_accuracy': 0.7200444270220828}
			------------EPOCH 25---------------
Loss:  tensor(60.2592, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(59.5172, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.8809, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.3369, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.2889, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(84.2356, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.5217, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.5529, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.5897, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.8791, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.5010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.9497, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.2281, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.0342, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.0092, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.7512, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.6780, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.0223, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.8411, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.7665, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.5697, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.3547, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.1973, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.1719, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.6641, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.6321, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.6955, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.4441, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.4233, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.1866, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.0380, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.4747, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.2110, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.2716, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.7832, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.7866, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.2755, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.9629, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(62.2648, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.2399, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.5205, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.7315, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.1825, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.6886, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.3250, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.6818, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.0845, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.3423, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.3859, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.0604, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.6460, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.1252, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.7715, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.4663, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.5392, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.8200, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.3079, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.6473, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.4331, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.2129032258064516, 'recall': 0.30697674418604654, 'f1': 0.25142857142857145, 'number': 215}, 'P': {'precision': 0.18341708542713567, 'recall': 0.27756653992395436, 'f1': 0.2208774583963691, 'number': 263}, 'overall_precision': 0.1963276836158192, 'overall_recall': 0.2907949790794979, 'overall_f1': 0.23440134907251264, 'overall_accuracy': 0.7142297138377107}
			------------EPOCH 26---------------
Loss:  tensor(50.7876, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.5408, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.0894, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.3850, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.2787, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(68.1080, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.0479, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.7091, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.9699, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.3661, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.1693, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.2709, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.0302, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.1751, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.8612, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.7884, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.9416, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.9876, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.6912, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.1127, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.8747, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.3918, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.3332, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.7458, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.8039, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.7583, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.1784, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.0301, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.0405, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.8992, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.5635, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.4848, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.3147, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.5098, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4628, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.6665, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.0759, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.1989, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.3940, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.4849, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.7640, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.3812, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.3070, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.8412, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.5956, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.4899, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.2919, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.4222, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.0176, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.8221, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.1135, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.4650, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.5905, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.0834, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.8009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.6351, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.7836, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.4877, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.7380, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.2145214521452145, 'recall': 0.3023255813953488, 'f1': 0.25096525096525096, 'number': 215}, 'P': {'precision': 0.16834170854271358, 'recall': 0.25475285171102663, 'f1': 0.2027231467473525, 'number': 263}, 'overall_precision': 0.18830242510699002, 'overall_recall': 0.27615062761506276, 'overall_f1': 0.22391857506361326, 'overall_accuracy': 0.7133803737096563}
			------------EPOCH 27---------------
Loss:  tensor(44.1640, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.1074, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.0662, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.4137, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.3911, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.6436, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.1150, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.8902, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.3319, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.0647, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.3537, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.0888, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.7623, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.7839, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.6667, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.0282, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.3386, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.1043, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.9790, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.5792, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.2544, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.5570, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.2193, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(62.3451, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.1589, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.8584, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.7606, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.1535, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.2247, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.4223, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.4779, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.9327, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.6120, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.4584, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2650, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2069, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.5620, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.6917, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.7363, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.6527, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.9846, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.3772, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.5612, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.3700, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.5125, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.5770, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.8498, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.0969, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.2946, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.9166, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.1611, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5807, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.2077, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.6885, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.0927, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.9513, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.4192, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.3333, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.6079, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.24149659863945577, 'recall': 0.3302325581395349, 'f1': 0.2789783889980354, 'number': 215}, 'P': {'precision': 0.17326732673267325, 'recall': 0.2661596958174905, 'f1': 0.2098950524737631, 'number': 263}, 'overall_precision': 0.2020057306590258, 'overall_recall': 0.29497907949790797, 'overall_f1': 0.23979591836734698, 'overall_accuracy': 0.7104403501894682}
			------------EPOCH 28---------------
Loss:  tensor(42.1185, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.6677, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.9196, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.7172, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.5674, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.0017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.4571, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.2447, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.8433, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.3797, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.0306, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.0034, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.1007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.4215, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.4492, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.1444, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.8555, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.5699, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.7658, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.1681, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.5247, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.0822, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.7554, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.8258, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.4681, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.0861, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.3659, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.2117, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.8802, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.6772, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.5990, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5540, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.0949, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.8881, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1303, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.8210, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.4549, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.9453, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.4869, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.9129, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.9860, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.3721, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.0121, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.3144, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.2905, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.8275, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.1801, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.6152, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.6732, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.3157, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.0568, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.3617, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.0338, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.5310, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.7583, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.2579, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.0701, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.4742, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.1610, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.21926910299003322, 'recall': 0.30697674418604654, 'f1': 0.2558139534883721, 'number': 215}, 'P': {'precision': 0.16377171215880892, 'recall': 0.2509505703422053, 'f1': 0.19819819819819817, 'number': 263}, 'overall_precision': 0.1875, 'overall_recall': 0.27615062761506276, 'overall_f1': 0.2233502538071066, 'overall_accuracy': 0.7129883705736313}
			------------EPOCH 29---------------
Loss:  tensor(55.7876, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.8510, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.5378, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.5100, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.9830, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.3253, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.6972, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.8552, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.0027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.9139, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.7501, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.3243, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.8245, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.8065, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.6731, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.4689, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.4193, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.3603, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.8874, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.2273, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.4727, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.2051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.5090, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.0086, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.3014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.1963, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.7569, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.6375, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.5389, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.3800, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.5723, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.1800, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.6838, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.0902, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9648, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5235, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.2576, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.1263, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.2031, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.9174, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.0277, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.8315, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.7849, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.6210, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5178, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5146, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.4218, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.0213, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.1943, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.9543, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2451, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.0573, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.4708, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.9849, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2069, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.6942, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.8631, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.6861, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.6553, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.2465753424657534, 'recall': 0.33488372093023255, 'f1': 0.2840236686390532, 'number': 215}, 'P': {'precision': 0.1708542713567839, 'recall': 0.2585551330798479, 'f1': 0.20574886535552192, 'number': 263}, 'overall_precision': 0.2028985507246377, 'overall_recall': 0.2928870292887029, 'overall_f1': 0.2397260273972603, 'overall_accuracy': 0.714360381549719}
			------------EPOCH 30---------------
Loss:  tensor(96.4877, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(64.1771, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(81.5179, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.7408, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.5995, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.7744, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.3932, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.4066, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.3262, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5388, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.4747, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.9423, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.4911, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.2723, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.3661, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.8812, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.8575, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.3527, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.3979, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.5852, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.9878, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.4300, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.0548, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.9337, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.6892, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.3070, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.2879, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.0435, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.9822, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.0799, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.7485, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.9945, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.4784, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.0109, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9249, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3966, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.4905, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.8787, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.9864, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.9024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.5684, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.1357, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.1153, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.7447, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.4162, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.1547, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.1475, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.8335, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.9273, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.0835, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2981, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.9223, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.3447, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.5449, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.9622, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.6752, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6884, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.1914, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.3635, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.2315112540192926, 'recall': 0.33488372093023255, 'f1': 0.2737642585551331, 'number': 215}, 'P': {'precision': 0.18181818181818182, 'recall': 0.2737642585551331, 'f1': 0.21851289833080423, 'number': 263}, 'overall_precision': 0.2036775106082037, 'overall_recall': 0.301255230125523, 'overall_f1': 0.2430379746835443, 'overall_accuracy': 0.7155363909577943}


		-------------RUN 5-----------
			------------EPOCH 1---------------
Loss:  tensor(1181.3286, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1214.7965, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3179.5420, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1687.3708, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3324.1631, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2176.6106, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2393.9248, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2719.5029, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1592.3918, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2315.4932, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1723.5793, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2112.4192, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2291.2046, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2079.5544, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2258.1147, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1635.9796, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2769.8716, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2297.0078, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2567.7151, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1430.7002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2807.2932, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2385.2695, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1504.7773, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1952.4509, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3221.2393, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2450.9946, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1116.4924, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(801.8021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1877.1741, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2329.6875, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1239.5085, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1616.1874, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1830.6204, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1767.7407, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1597.4958, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1614.8961, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2151.0239, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1331.9647, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1500.8978, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1991.5066, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1996.3361, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1938.0137, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1721.8259, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1378.8240, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1498.5798, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1980.1870, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1222.5347, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1023.2296, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(900.8661, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1557.2126, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2201.6731, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1452.6658, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1156.6322, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1422.7797, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1202.2416, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1865.3176, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.030864197530864196, 'recall': 0.014245014245014245, 'f1': 0.01949317738791423, 'number': 351}, 'P': {'precision': 0.05855855855855856, 'recall': 0.0912280701754386, 'f1': 0.07133058984910835, 'number': 285}, 'overall_precision': 0.05115511551155116, 'overall_recall': 0.04874213836477988, 'overall_f1': 0.049919484702093404, 'overall_accuracy': 0.567979669631512}
			------------EPOCH 2---------------
Loss:  tensor(715.0057, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(671.2210, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2077.2971, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(931.5200, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1798.6016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1500.5244, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1721.1101, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2092.5518, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1168.7205, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1567.7848, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1139.3586, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1460.3005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1812.4954, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1677.5548, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1546.7341, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1234.2059, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2581.3323, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2030.0326, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2036.1121, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1161.6218, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2381.3672, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2019.6255, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1159.6471, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1686.4670, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2868.4512, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2102.0166, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(966.7828, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(646.1095, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1613.5439, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2042.5032, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1110.0248, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1428.0668, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1710.6914, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1510.4045, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1346.1110, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1425.0330, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1934.4200, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1112.3446, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1291.3447, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1779.7257, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1604.2188, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1502.8967, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1533.3301, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1202.1903, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1228.4497, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1654.1228, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1040.4490, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(877.1652, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(798.6807, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1315.4993, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1966.6509, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1281.1492, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(981.2467, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1232.7853, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(998.8013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1615.9093, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.10508474576271186, 'recall': 0.08831908831908832, 'f1': 0.09597523219814241, 'number': 351}, 'P': {'precision': 0.16083916083916083, 'recall': 0.24210526315789474, 'f1': 0.19327731092436976, 'number': 285}, 'overall_precision': 0.13812154696132597, 'overall_recall': 0.15723270440251572, 'overall_f1': 0.14705882352941177, 'overall_accuracy': 0.6101320369040385}
			------------EPOCH 3---------------
Loss:  tensor(578.3113, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(563.3088, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1841.1277, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(750.0447, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1364.5347, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1223.4296, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1447.2913, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1805.6595, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(988.9767, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1328.3440, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(873.1780, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1227.8361, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1561.8633, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1511.7686, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1263.1018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1020.8132, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2157.5596, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1667.8879, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1724.3984, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(931.0490, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1992.4219, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1749.1370, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(978.8634, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1472.5300, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2547.8608, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1783.9526, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(839.1150, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(420.1745, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1326.8961, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1629.2155, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(887.6912, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1107.5626, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1340.0331, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1325.6553, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1216.6801, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1301.3965, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1622.2518, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1015.2587, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1113.4485, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1659.0768, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1254.7242, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1215.0352, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1210.1572, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1074.3146, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(996.3468, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1386.4827, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(872.2223, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(724.2941, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(650.0803, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(937.7764, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1568.4927, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1088.6462, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(886.8050, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1145.3751, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(987.4753, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1497.1721, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.19170984455958548, 'recall': 0.10541310541310542, 'f1': 0.13602941176470587, 'number': 351}, 'P': {'precision': 0.21350762527233116, 'recall': 0.34385964912280703, 'f1': 0.26344086021505375, 'number': 285}, 'overall_precision': 0.20705521472392638, 'overall_recall': 0.21226415094339623, 'overall_f1': 0.20962732919254656, 'overall_accuracy': 0.6170377327219491}
			------------EPOCH 4---------------
Loss:  tensor(434.9456, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(496.3013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1779.3254, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(644.6568, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1015.4387, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1044.7629, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1216.4512, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1581.5823, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(876.9106, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1128.8621, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(662.5254, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(995.2161, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1326.0107, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1159.4229, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1042.2849, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(833.8450, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1701.7292, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1369.4827, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1541.0378, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(750.7444, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1745.4995, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1517.2594, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(861.2338, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1232.2451, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2320.8306, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1482.2382, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(702.6617, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(339.6111, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1108.4003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1182.3792, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(646.4464, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(751.5814, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(968.1971, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1065.3296, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(976.4380, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1148.3086, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1432.4243, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(752.3224, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(918.1922, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1326.6724, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(902.5468, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(905.8461, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(855.3757, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(863.6215, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(843.5062, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1173.9838, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(659.3580, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(531.6447, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(485.4326, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(748.6177, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1312.8127, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(930.5642, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(706.7268, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(925.2236, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(781.9557, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1211.8499, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.23414634146341465, 'recall': 0.13675213675213677, 'f1': 0.17266187050359713, 'number': 351}, 'P': {'precision': 0.16427104722792607, 'recall': 0.2807017543859649, 'f1': 0.20725388601036268, 'number': 285}, 'overall_precision': 0.18497109826589594, 'overall_recall': 0.20125786163522014, 'overall_f1': 0.1927710843373494, 'overall_accuracy': 0.6320092812551793}
			------------EPOCH 5---------------
Loss:  tensor(334.9736, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(408.2604, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1343.0742, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(543.8578, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(867.2335, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(862.3849, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1021.7853, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1274.5411, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(744.9346, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(965.2518, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(404.4857, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(760.3026, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1046.1914, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(905.6255, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(839.2440, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(651.9274, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1277.9418, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1086.5458, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1403.9048, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(589.7245, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1431.8396, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1216.7062, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(626.7303, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(939.9888, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1820.4866, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(970.9962, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(568.4812, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(219.0989, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(875.1418, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(839.4484, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(422.9375, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(393.0998, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(649.6241, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(915.9596, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(798.6342, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(965.1364, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1197.5718, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(599.1351, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(764.3223, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1151.6294, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(654.0834, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(623.6483, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(560.9911, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(630.7501, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(723.2910, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(931.0857, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(415.1904, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(302.1349, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(344.8946, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(595.2618, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1083.1515, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(779.2651, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(552.2645, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(749.4915, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(573.5945, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(877.9310, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.29012345679012347, 'recall': 0.1339031339031339, 'f1': 0.18323586744639378, 'number': 351}, 'P': {'precision': 0.15316901408450703, 'recall': 0.30526315789473685, 'f1': 0.2039859320046893, 'number': 285}, 'overall_precision': 0.18356164383561643, 'overall_recall': 0.21069182389937108, 'overall_f1': 0.19619326500732065, 'overall_accuracy': 0.625545549969615}
			------------EPOCH 6---------------
Loss:  tensor(242.5793, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(360.2533, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1012.6249, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(444.2113, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(662.8550, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(808.2498, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(848.0316, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(960.5416, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(713.2939, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(917.6694, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(338.8207, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(690.6505, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(788.6452, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(801.2228, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(722.6068, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(592.3673, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(842.8993, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(781.1891, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1136.1014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(409.4729, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1170.7682, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(933.9325, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(474.5379, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(733.4790, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1528.9067, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(748.9830, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(516.7426, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(174.5916, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(880.5269, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(797.2817, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(318.7376, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(305.4722, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(359.1165, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(811.5919, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(655.1321, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(889.9152, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(865.7482, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(443.8117, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(582.5822, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(919.4834, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(656.1390, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(655.4594, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(551.7694, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(599.6152, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(722.3860, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(963.7891, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(511.2695, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(320.1562, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(302.2264, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(545.9472, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(971.6005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(576.0833, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(364.0167, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(555.1401, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(452.5901, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(631.9282, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.27124183006535946, 'recall': 0.23646723646723647, 'f1': 0.2526636225266362, 'number': 351}, 'P': {'precision': 0.2545454545454545, 'recall': 0.3929824561403509, 'f1': 0.30896551724137933, 'number': 285}, 'overall_precision': 0.2613941018766756, 'overall_recall': 0.30660377358490565, 'overall_f1': 0.2821997105643994, 'overall_accuracy': 0.6289707750952986}
			------------EPOCH 7---------------
Loss:  tensor(222.0484, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(326.0195, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1058.1802, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(455.9947, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(593.0760, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(612.7201, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(795.2773, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1227.7134, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(638.8458, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(767.1069, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(289.9993, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(613.5758, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(886.3684, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(762.2542, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(618.3744, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(505.0005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(891.8123, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(879.7427, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1157.3480, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(476.2086, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1125.6759, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(888.2869, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(447.6841, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(699.3914, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1307.9626, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(586.2610, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(287.2997, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(139.2891, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(692.7627, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(604.0511, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(245.5280, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(221.1158, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(300.1613, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(738.9053, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(574.5027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(715.3834, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1002.8849, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(708.9127, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(735.8342, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1189.0566, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(763.6531, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(650.2859, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(593.1776, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(449.5727, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(657.7086, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(704.7799, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(335.6335, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(202.9133, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(320.0560, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(601.0956, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1225.6459, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(708.5013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(491.9545, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(771.2441, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(572.2517, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(822.8073, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.25, 'recall': 0.10826210826210826, 'f1': 0.1510934393638171, 'number': 351}, 'P': {'precision': 0.11233480176211454, 'recall': 0.17894736842105263, 'f1': 0.13802435723951287, 'number': 285}, 'overall_precision': 0.14686468646864687, 'overall_recall': 0.139937106918239, 'overall_f1': 0.14331723027375204, 'overall_accuracy': 0.6105187558698415}
			------------EPOCH 8---------------
Loss:  tensor(218.7924, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(315.1528, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(902.1911, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(271.7188, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(516.9630, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(631.2268, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(713.5038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(896.6659, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(512.3207, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(674.6669, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(211.3975, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(501.6122, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(610.2917, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(672.4868, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(546.5183, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(431.2904, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(744.0465, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(682.8799, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(924.6578, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(418.6314, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1149.6946, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(849.5219, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(419.3571, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(715.8182, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1540.4139, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(696.1686, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(371.4492, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(183.4723, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(933.8564, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1188.2866, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(410.3142, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(286.3821, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(389.6570, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(645.4092, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(575.9186, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(566.7964, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1182.5281, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(519.2189, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(511.6013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(950.5128, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(581.6674, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(445.5409, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(474.4128, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(411.2850, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(582.1381, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(721.8277, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(304.9298, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(184.7048, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(260.9294, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(556.8450, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1071.8271, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(560.6399, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(353.6574, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(531.0919, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(481.6991, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(751.9431, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.13793103448275862, 'recall': 0.06837606837606838, 'f1': 0.09142857142857144, 'number': 351}, 'P': {'precision': 0.04375, 'recall': 0.02456140350877193, 'f1': 0.031460674157303366, 'number': 285}, 'overall_precision': 0.09281437125748503, 'overall_recall': 0.04874213836477988, 'overall_f1': 0.06391752577319587, 'overall_accuracy': 0.483232970554113}
			------------EPOCH 9---------------
Loss:  tensor(442.2512, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(325.5247, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1510.7460, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(731.9745, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1838.2805, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(965.4978, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1167.2551, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1731.1616, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(683.5176, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(928.5585, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(236.1153, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(604.1005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(593.1693, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(605.1929, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(477.8661, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(427.9677, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(495.5851, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(460.4211, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(617.9843, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(245.4404, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(857.2527, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(701.9986, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(361.1286, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(533.6447, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1256.3857, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(553.2169, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(300.2026, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(151.5693, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(897.0379, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1136.9221, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(545.5457, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(303.4507, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(437.5016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(652.5809, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(642.2192, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(743.5164, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1411.1206, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(565.7449, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(662.1139, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(906.6335, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(589.0249, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(434.6742, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(666.0082, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(642.2809, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(521.5097, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(667.6947, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(449.7073, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(281.5284, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(329.6366, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(792.7487, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1118.9194, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(729.0314, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(583.3920, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(806.7994, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(531.2386, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(679.8327, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.1509433962264151, 'recall': 0.25071225071225073, 'f1': 0.18843683083511778, 'number': 351}, 'P': {'precision': 0.06153846153846154, 'recall': 0.07017543859649122, 'f1': 0.06557377049180328, 'number': 285}, 'overall_precision': 0.11894273127753303, 'overall_recall': 0.16981132075471697, 'overall_f1': 0.13989637305699482, 'overall_accuracy': 0.636981382244075}
			------------EPOCH 10---------------
Loss:  tensor(224.5260, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(296.4815, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(764.5385, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(465.0490, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(811.8271, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(536.7342, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(506.0315, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(833.9245, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(426.0499, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(663.0325, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(222.2181, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(422.8647, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(619.6628, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(581.1792, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(509.6021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(448.3672, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(854.3131, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(790.1520, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(966.7515, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(412.0647, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1122.2411, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(932.1617, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(560.7947, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(661.7106, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1473.3544, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(777.3811, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(343.7563, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(225.7691, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(599.8068, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(507.0117, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(459.8344, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(252.8761, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(329.6965, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(684.3403, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(667.1458, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(535.4125, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(970.2548, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(269.1925, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(574.4868, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(514.4674, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(374.0820, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(366.7735, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(325.8580, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(302.9321, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(366.1159, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(544.0186, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(221.8716, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(125.9102, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(169.4097, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(467.0213, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(597.4040, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(343.5429, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(307.8214, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(527.7949, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(378.4205, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(497.1315, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.2182468694096601, 'recall': 0.3475783475783476, 'f1': 0.2681318681318681, 'number': 351}, 'P': {'precision': 0.2222222222222222, 'recall': 0.06315789473684211, 'f1': 0.09836065573770493, 'number': 285}, 'overall_precision': 0.21875, 'overall_recall': 0.22012578616352202, 'overall_f1': 0.219435736677116, 'overall_accuracy': 0.5977017844317993}
			------------EPOCH 11---------------
Loss:  tensor(243.6554, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(269.3460, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1068.2889, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(434.9933, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1059.9534, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(731.4599, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(699.7490, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1453.4275, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(436.2421, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(874.6472, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(319.8266, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(389.9153, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(814.5992, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(623.5785, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(557.9763, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(420.8590, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(525.7087, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(318.8615, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(563.5295, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(167.1406, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(672.4503, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(544.7179, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(277.5637, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(388.4165, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(808.3491, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(346.9306, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(180.5797, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(91.2411, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(444.7101, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(371.3744, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(328.1172, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(165.6660, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(489.4740, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(656.0759, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(683.2123, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(611.1255, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1108.0630, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(236.6723, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(871.8038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(692.4839, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(502.4324, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(420.1528, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(429.1880, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(507.3918, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(597.2538, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(699.9328, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(387.5504, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(171.7408, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(183.5995, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(471.8035, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(742.4442, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(385.8224, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(258.8472, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(423.8083, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(367.1490, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(382.5522, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.2, 'recall': 0.23076923076923078, 'f1': 0.21428571428571427, 'number': 351}, 'P': {'precision': 0.05573770491803279, 'recall': 0.05964912280701754, 'f1': 0.0576271186440678, 'number': 285}, 'overall_precision': 0.13802816901408452, 'overall_recall': 0.1540880503144654, 'overall_f1': 0.14561664190193166, 'overall_accuracy': 0.6187503452847909}
			------------EPOCH 12---------------
Loss:  tensor(136.1389, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(147.8503, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(390.3030, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(221.6759, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(473.0876, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(344.2820, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(371.6040, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(572.8215, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(277.0209, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(521.9709, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(181.5755, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(291.3635, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(672.6562, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(460.3216, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(396.5413, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(349.7757, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(750.0328, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(575.7380, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(880.2439, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(282.7554, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(873.2380, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(802.0936, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(380.4810, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(512.4856, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(929.5121, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(437.7694, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(243.3971, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(78.9655, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(454.9268, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(327.3086, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(131.6946, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(106.8265, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(151.7587, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(376.6672, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(340.3345, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(264.7144, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(405.2445, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(153.2431, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(236.4521, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(280.0050, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(210.7518, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(204.8962, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(180.8123, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(136.6141, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(227.2504, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(299.4109, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(136.0155, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(75.4332, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(100.9277, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(288.8773, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(411.6538, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(208.0604, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(188.4203, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(378.5564, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(270.1318, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(310.0615, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.34080717488789236, 'recall': 0.21652421652421652, 'f1': 0.26480836236933797, 'number': 351}, 'P': {'precision': 0.23015873015873015, 'recall': 0.4070175438596491, 'f1': 0.29404309252217997, 'number': 285}, 'overall_precision': 0.2640990371389271, 'overall_recall': 0.3018867924528302, 'overall_f1': 0.28173147468818777, 'overall_accuracy': 0.6702392133031324}
			------------EPOCH 13---------------
Loss:  tensor(97.1175, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(184.7023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(453.9240, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(232.6237, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(315.2317, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(330.4885, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(391.3088, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(505.0213, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(369.4783, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(468.9765, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(127.6053, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(229.1426, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(359.8264, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(361.7541, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(317.3611, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(224.6310, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(259.8381, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(224.6635, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(297.0864, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(115.1723, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(529.5534, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(367.0963, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(165.0116, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(279.5642, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(509.2728, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(189.8833, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(74.0112, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.1706, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(297.8173, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(225.5917, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(102.9696, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(67.2765, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(116.2415, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(306.2444, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(263.2419, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(179.4417, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(323.4781, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(134.6740, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(185.3288, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(239.2256, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(204.9534, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(125.8346, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(167.4105, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(76.6115, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(170.8061, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(249.8587, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(123.9312, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.5453, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(107.7124, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(229.2399, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(312.2405, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(161.2835, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(98.2150, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(200.3570, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(163.8722, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(169.7728, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.2700892857142857, 'recall': 0.34472934472934474, 'f1': 0.3028785982478097, 'number': 351}, 'P': {'precision': 0.1554054054054054, 'recall': 0.16140350877192983, 'f1': 0.15834767641996556, 'number': 285}, 'overall_precision': 0.22446236559139784, 'overall_recall': 0.26257861635220126, 'overall_f1': 0.24202898550724636, 'overall_accuracy': 0.6743826307938788}
			------------EPOCH 14---------------
Loss:  tensor(77.0059, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(98.0045, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(238.0570, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(108.4359, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(244.6987, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(170.2160, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(191.3194, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(291.8649, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(172.8533, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(263.6649, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(84.2189, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(104.5612, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(283.1131, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(239.0637, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(150.0942, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(147.4709, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(148.3185, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(98.9164, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(225.6811, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(70.5835, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(385.1669, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(233.1496, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(125.0755, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(181.9281, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(433.6951, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(164.3689, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.2457, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.4927, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(249.7105, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(175.3490, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(78.8162, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.2845, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(74.4052, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(230.5078, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(214.9861, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(130.0126, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(260.1657, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(79.3776, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(131.1595, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(153.6634, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(103.7430, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(98.7526, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(124.0730, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.6029, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(134.8777, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(168.7186, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(74.2789, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.1538, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.4112, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(158.3478, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(197.0429, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(110.6937, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(77.2128, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(142.9817, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(124.0360, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(124.2273, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.2735294117647059, 'recall': 0.26495726495726496, 'f1': 0.2691751085383502, 'number': 351}, 'P': {'precision': 0.18627450980392157, 'recall': 0.26666666666666666, 'f1': 0.21933621933621936, 'number': 285}, 'overall_precision': 0.22593582887700533, 'overall_recall': 0.26572327044025157, 'overall_f1': 0.24421965317919075, 'overall_accuracy': 0.6601845201922546}
			------------EPOCH 15---------------
Loss:  tensor(52.1457, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(84.3248, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(183.3143, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(82.0093, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(138.4679, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(95.6253, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(138.0507, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(206.8064, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(148.4192, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(202.1294, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.7340, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(69.3859, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(192.2144, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(178.7919, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(102.5066, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(98.8589, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(103.1079, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(72.8387, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(178.8218, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.7562, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(316.3103, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(148.4212, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(69.7416, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(132.9277, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(328.1170, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(116.7685, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.2354, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.7100, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(203.6142, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(135.3193, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(61.4472, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.4185, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.2009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(183.1404, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(153.0948, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(94.1532, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(206.2771, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(66.5690, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(102.5678, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(120.7879, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(83.1584, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.3625, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(90.0962, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.3500, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(87.0438, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(109.3964, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.2082, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.7116, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.2969, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(116.3176, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(140.5078, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(87.9491, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.4016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(98.2278, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(92.7001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(83.2227, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.2734584450402145, 'recall': 0.2905982905982906, 'f1': 0.281767955801105, 'number': 351}, 'P': {'precision': 0.1853932584269663, 'recall': 0.23157894736842105, 'f1': 0.2059282371294852, 'number': 285}, 'overall_precision': 0.23045267489711935, 'overall_recall': 0.2641509433962264, 'overall_f1': 0.24615384615384614, 'overall_accuracy': 0.6726147726644937}
			------------EPOCH 16---------------
Loss:  tensor(41.7635, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(64.3361, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(121.5494, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.1228, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(102.5901, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(64.7415, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(103.0367, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(174.6233, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(113.3404, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(148.9141, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.4483, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.6382, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(151.3647, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(139.4150, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.5227, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(61.8524, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(73.9163, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.6260, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(145.1719, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.1962, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(266.2090, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(116.9780, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.6963, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(103.8890, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(261.5795, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(94.3769, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.1314, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.8693, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(156.1542, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(104.0712, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.6000, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.6984, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.0075, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(149.8237, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(123.4937, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(67.3137, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(171.0251, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.6943, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(73.6482, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(95.5549, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.8703, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.1726, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(74.3938, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.9112, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(69.7558, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(79.1223, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.6443, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.7186, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.9195, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(100.5707, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(110.4055, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(69.7095, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.2896, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(69.4335, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(75.1300, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.0523, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.2711864406779661, 'recall': 0.27350427350427353, 'f1': 0.2723404255319149, 'number': 351}, 'P': {'precision': 0.16030534351145037, 'recall': 0.22105263157894736, 'f1': 0.18584070796460178, 'number': 285}, 'overall_precision': 0.21285140562248997, 'overall_recall': 0.25, 'overall_f1': 0.22993492407809113, 'overall_accuracy': 0.6652671123142367}
			------------EPOCH 17---------------
Loss:  tensor(30.2097, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.4815, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(88.9441, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.6890, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(89.2520, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.6416, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(79.3417, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(141.5871, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(98.2206, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(121.3007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.6844, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.5701, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(126.3029, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(114.9310, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.6847, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.9764, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.6512, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.5518, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(124.6549, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.8904, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(231.5808, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(88.3950, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.1194, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(77.7855, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(203.6208, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(70.9441, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.5664, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.4897, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(133.0411, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(90.9953, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.3142, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.9743, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.2525, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(120.1196, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(100.8489, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.3966, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(140.9316, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.9422, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.2597, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(69.7322, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.5486, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.5993, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(65.6468, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.0072, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.4564, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.2236, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.2484, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.5754, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.4040, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(74.5604, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(79.5217, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.1089, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.2018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.8746, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(75.2284, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(62.9866, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.29198966408268734, 'recall': 0.32193732193732194, 'f1': 0.3062330623306233, 'number': 351}, 'P': {'precision': 0.18873239436619718, 'recall': 0.23508771929824562, 'f1': 0.209375, 'number': 285}, 'overall_precision': 0.24258760107816713, 'overall_recall': 0.2830188679245283, 'overall_f1': 0.26124818577648773, 'overall_accuracy': 0.6746036130600519}
			------------EPOCH 18---------------
Loss:  tensor(21.9569, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.1387, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(68.5185, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.9570, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(75.3400, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.1898, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(59.9614, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(121.2850, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(77.8632, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(97.2726, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.8043, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.2803, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(105.8948, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(105.1101, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.1147, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.5300, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.7555, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.1016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(108.2431, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.3886, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(200.9602, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(69.8922, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.2569, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(61.0078, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(170.6518, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.0162, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.1187, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.7301, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(116.0723, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(79.9498, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.1903, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.5399, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.5178, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(102.4962, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(86.2531, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.5941, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(119.2442, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.9081, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.5402, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.6396, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.1873, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.1964, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.1265, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.8914, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.0190, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.4698, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.9839, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.0585, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.7184, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.1152, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.6648, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.6547, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.0425, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.9909, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(70.0279, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.8267, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.2770083102493075, 'recall': 0.2849002849002849, 'f1': 0.28089887640449446, 'number': 351}, 'P': {'precision': 0.19638242894056848, 'recall': 0.26666666666666666, 'f1': 0.2261904761904762, 'number': 285}, 'overall_precision': 0.23529411764705882, 'overall_recall': 0.27672955974842767, 'overall_f1': 0.2543352601156069, 'overall_accuracy': 0.6623390972874427}
			------------EPOCH 19---------------
Loss:  tensor(15.9504, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.8619, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.0362, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.3472, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(68.8327, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.5918, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.4130, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(96.9446, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(64.5298, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(80.7658, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.0692, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.9086, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(86.1578, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(84.7022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.2748, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.2694, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.2730, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.8096, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(95.8595, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.4506, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(174.4683, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.8558, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.3794, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.3850, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(161.2911, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.1068, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.5642, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.5493, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(106.3527, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(79.6363, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.7561, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.2302, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.8572, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(78.3545, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(80.8055, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.8986, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(95.5354, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.5942, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.1586, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.4888, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.7059, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.9098, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.8843, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.0043, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.6348, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.7503, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.4293, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.6502, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.0182, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.9589, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.6281, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.5191, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.5391, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.7367, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.2656, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.4579, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.27692307692307694, 'recall': 0.3076923076923077, 'f1': 0.29149797570850206, 'number': 351}, 'P': {'precision': 0.2062146892655367, 'recall': 0.256140350877193, 'f1': 0.22848200312989045, 'number': 285}, 'overall_precision': 0.24327956989247312, 'overall_recall': 0.28459119496855345, 'overall_f1': 0.2623188405797101, 'overall_accuracy': 0.6751560687254847}
			------------EPOCH 20---------------
Loss:  tensor(12.7722, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.6474, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.7210, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.9540, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(64.2398, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.6743, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.7935, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(95.2670, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.2826, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(74.0743, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.5485, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.9782, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(71.5679, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(85.0475, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.4482, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.1665, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.1085, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.3982, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(90.3351, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.7987, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(154.1052, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.4832, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.4257, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.9592, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(132.3791, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.6224, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.7675, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.9474, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(79.2402, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.7594, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.9525, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.7128, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.2416, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(66.4932, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(84.7259, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.1402, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(84.2304, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.8512, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(66.1113, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.3871, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.4923, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.2177, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.2757, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.1704, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.1396, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.7587, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.5022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.5798, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.3931, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.5658, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.5149, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.6254, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.2683, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.7700, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.5357, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.0753, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.2757660167130919, 'recall': 0.28205128205128205, 'f1': 0.2788732394366198, 'number': 351}, 'P': {'precision': 0.1815856777493606, 'recall': 0.24912280701754386, 'f1': 0.2100591715976331, 'number': 285}, 'overall_precision': 0.22666666666666666, 'overall_recall': 0.2672955974842767, 'overall_f1': 0.2453102453102453, 'overall_accuracy': 0.6639412187171979}
			------------EPOCH 21---------------
Loss:  tensor(10.1584, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.6953, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.9856, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.4206, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(65.6285, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.7580, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.9826, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(71.6147, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.3442, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.7054, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.5130, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.6355, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(65.5162, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.2674, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.0342, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.9631, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.0239, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.7085, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(82.0817, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.1700, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(135.0657, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.5656, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.0117, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.8946, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(113.9204, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.5566, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.4690, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.6284, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(68.4805, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.7889, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.4280, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.1898, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.4431, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.6513, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.2096, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.8024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(69.7528, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.4328, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.4733, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(67.5697, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.1225, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.1545, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(67.7121, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.7091, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.1872, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.8470, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.6794, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.1023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.9502, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.0284, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.3217, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.5671, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.3781, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.5271, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.0089, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.6421, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.27835051546391754, 'recall': 0.3076923076923077, 'f1': 0.2922868741542625, 'number': 351}, 'P': {'precision': 0.211864406779661, 'recall': 0.2631578947368421, 'f1': 0.23474178403755866, 'number': 285}, 'overall_precision': 0.24663072776280323, 'overall_recall': 0.28773584905660377, 'overall_f1': 0.26560232220609575, 'overall_accuracy': 0.6809016076459864}
			------------EPOCH 22---------------
Loss:  tensor(8.4421, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.0220, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.3605, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.0501, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.2242, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.5041, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.3271, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.3293, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.9912, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.3757, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.6539, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.3116, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.9250, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.8735, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.2359, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.7134, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.9465, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.6119, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(68.3592, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.7812, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(123.2912, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.3672, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.9107, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.4146, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(87.5166, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.8719, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.2043, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5103, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.8097, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.9219, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.6979, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.8701, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.9341, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.2090, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.5511, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.7240, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.3563, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.5407, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.8006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.5902, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.9407, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.0227, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.9145, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.7798, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.1790, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.4018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.7046, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.4150, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.2142, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.3201, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.0215, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.8264, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.0035, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.2894, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.4740, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.3884, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.2773109243697479, 'recall': 0.28205128205128205, 'f1': 0.2796610169491525, 'number': 351}, 'P': {'precision': 0.17258883248730963, 'recall': 0.23859649122807017, 'f1': 0.20029455081001474, 'number': 285}, 'overall_precision': 0.22237017310252996, 'overall_recall': 0.26257861635220126, 'overall_f1': 0.24080749819754868, 'overall_accuracy': 0.6747141041931385}
			------------EPOCH 23---------------
Loss:  tensor(7.1162, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.0883, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.1726, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.6038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.8236, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.4748, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.8718, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.3512, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.9992, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.6299, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.3578, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.3005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.9454, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.1813, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.5758, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.0206, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.8463, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.0558, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.6679, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.6623, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(109.6954, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.5316, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.2923, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.0115, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(67.3922, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.9425, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.6354, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2564, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.9671, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.4070, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.1999, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.2579, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.5465, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.9885, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.3168, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.8925, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.0895, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.6899, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.4206, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.9122, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.9584, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.8274, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.7790, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.0311, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.0319, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.3240, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.7661, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.9267, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.1555, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.3209, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.4773, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.4473, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.6226, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.8758, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.3106, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.7913, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.2868217054263566, 'recall': 0.3162393162393162, 'f1': 0.30081300813008127, 'number': 351}, 'P': {'precision': 0.21052631578947367, 'recall': 0.26666666666666666, 'f1': 0.23529411764705882, 'number': 285}, 'overall_precision': 0.25, 'overall_recall': 0.2940251572327044, 'overall_f1': 0.2702312138728324, 'overall_accuracy': 0.683166675874261}
			------------EPOCH 24---------------
Loss:  tensor(5.7811, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.4137, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.4759, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.5548, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.9890, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.9070, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.0044, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.5450, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.5435, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.3829, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.7522, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.7516, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.1585, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.1574, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.0325, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.5764, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.1408, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.9419, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.9710, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.6378, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(83.7962, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.6368, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.7791, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.8004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(80.9928, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.9061, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.0054, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.7612, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.2250, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.2238, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.0725, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.8263, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.9155, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.8705, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.6552, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.3918, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.0718, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.0054, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.6785, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.2852, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.3681, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.4443, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.7369, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.4520, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.7803, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.0717, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.0235, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3782, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.0443, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.4626, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.0333, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.8621, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.3434, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.9044, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.9353, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.0246, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.2707774798927614, 'recall': 0.28774928774928776, 'f1': 0.27900552486187846, 'number': 351}, 'P': {'precision': 0.1834625322997416, 'recall': 0.24912280701754386, 'f1': 0.2113095238095238, 'number': 285}, 'overall_precision': 0.22631578947368422, 'overall_recall': 0.27044025157232704, 'overall_f1': 0.24641833810888253, 'overall_accuracy': 0.6760952433567207}
			------------EPOCH 25---------------
Loss:  tensor(5.4063, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.2857, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.0603, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.0887, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.1365, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.1317, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.5039, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.4166, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.6242, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.6566, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.0297, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.1050, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.8712, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.3424, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.8176, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.5926, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.7348, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.1082, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.2977, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.7037, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(64.5169, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(66.0575, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.2815, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.4621, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(95.9750, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.2928, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5367, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5555, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.2401, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.2597, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.5756, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.0108, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.3191, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.4818, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.6570, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.2930, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.6413, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.0170, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.3762, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.6591, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.3041, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.9481, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.0153, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.1591, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.6100, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.3319, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.7933, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.8182, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.5388, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.7242, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.6059, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.6620, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.7786, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.6751, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.8364, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.5158, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.2826086956521739, 'recall': 0.2962962962962963, 'f1': 0.2892906815020862, 'number': 351}, 'P': {'precision': 0.18925831202046037, 'recall': 0.2596491228070175, 'f1': 0.21893491124260356, 'number': 285}, 'overall_precision': 0.23451910408432147, 'overall_recall': 0.279874213836478, 'overall_f1': 0.2551971326164874, 'overall_accuracy': 0.6768134357217833}
			------------EPOCH 26---------------
Loss:  tensor(4.7315, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.0139, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.1640, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.5077, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.0098, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.4774, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.3888, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.0913, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.5274, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.3411, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.4631, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.5450, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.1313, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.6433, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.8075, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.8984, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.6720, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.5262, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(62.7120, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.2760, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(64.0654, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(114.5982, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.1840, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.4456, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(79.9811, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.7020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.4382, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.7429, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.5869, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.0611, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.3777, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.3638, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.3171, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.5461, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.9309, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.6203, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.8010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.3618, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.3933, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.6413, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.8400, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.2521, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.5513, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.9656, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.6170, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.3037, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.3785, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9835, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5223, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.4578, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.8012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.9193, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.0846, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.4960, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.3673, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.3229, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.275609756097561, 'recall': 0.32193732193732194, 'f1': 0.29697766097240474, 'number': 351}, 'P': {'precision': 0.18882978723404256, 'recall': 0.24912280701754386, 'f1': 0.21482602118003027, 'number': 285}, 'overall_precision': 0.2340966921119593, 'overall_recall': 0.2893081761006289, 'overall_f1': 0.25879043600562585, 'overall_accuracy': 0.6807358709463566}
			------------EPOCH 27---------------
Loss:  tensor(4.3325, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.0553, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.7218, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.9227, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.8629, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.1833, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.5037, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.8941, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.2013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.9608, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.6212, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.1971, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.9564, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.3200, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.5856, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.7109, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.6651, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.9132, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.6702, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.0412, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(83.0295, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(113.1937, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.0754, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.2874, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(92.9189, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.0932, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.0540, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3675, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.9828, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.9059, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.5934, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.8906, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.8880, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.7486, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.6566, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.1858, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.2905, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.4705, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.4933, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.9422, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.1811, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.3163, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.9253, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5496, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.0914, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.1598, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.1366, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9517, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.1804, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.9430, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.4104, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.9988, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.5425, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.5623, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.3225, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.8873, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.2782608695652174, 'recall': 0.27350427350427353, 'f1': 0.2758620689655172, 'number': 351}, 'P': {'precision': 0.1681614349775785, 'recall': 0.2631578947368421, 'f1': 0.20519835841313272, 'number': 285}, 'overall_precision': 0.21618204804045513, 'overall_recall': 0.2688679245283019, 'overall_f1': 0.23966362999299226, 'overall_accuracy': 0.6712888790674548}
			------------EPOCH 28---------------
Loss:  tensor(4.1807, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.9182, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.3061, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.0876, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.5550, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.2073, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.6531, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.2974, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.8418, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(78.2744, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.8051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.1444, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.7617, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(69.7520, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.2698, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.0751, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.1326, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.8029, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(70.4365, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.9278, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(64.5862, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(74.1153, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.8036, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.3146, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(83.7631, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.7786, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.9503, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1145, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.3405, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.0797, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.2561, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.4918, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.3673, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.0876, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.3706, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.8347, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.5185, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.8599, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.3807, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.5529, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.4738, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.0549, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.3088, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.7474, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.4053, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.8701, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2350, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1226, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.1924, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.8464, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.0556, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.0357, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.5438, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.1775, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.9156, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.6786, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.2736318407960199, 'recall': 0.31339031339031337, 'f1': 0.2921646746347942, 'number': 351}, 'P': {'precision': 0.20821114369501467, 'recall': 0.24912280701754386, 'f1': 0.2268370607028754, 'number': 285}, 'overall_precision': 0.24360699865410498, 'overall_recall': 0.28459119496855345, 'overall_f1': 0.26250906453952144, 'overall_accuracy': 0.6818407822772222}
			------------EPOCH 29---------------
Loss:  tensor(3.6391, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.9600, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.0400, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.7707, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.4008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.6994, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.8127, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.2156, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.3779, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(119.6762, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.8354, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.6992, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.2929, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(89.4738, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.2319, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.9754, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.2022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.4586, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.2458, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.6419, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(76.2035, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(106.5977, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.0697, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.2488, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.1928, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.4312, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.0883, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6549, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(84.2207, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.5402, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.7945, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.4981, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.8241, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.1100, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.3426, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.0252, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.5592, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.1282, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.0930, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.6998, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.5557, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.8207, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.8648, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.4282, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.8632, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.4629, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2548, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8863, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.9317, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.8977, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.5971, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.9400, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.2215, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.7507, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.2411, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.7390, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.2830188679245283, 'recall': 0.29914529914529914, 'f1': 0.29085872576177285, 'number': 351}, 'P': {'precision': 0.1744186046511628, 'recall': 0.2631578947368421, 'f1': 0.20979020979020976, 'number': 285}, 'overall_precision': 0.2247191011235955, 'overall_recall': 0.2830188679245283, 'overall_f1': 0.25052192066805845, 'overall_accuracy': 0.6746588586265952}
			------------EPOCH 30---------------
Loss:  tensor(3.4873, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.5295, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.5388, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.8187, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.9578, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.7069, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.4392, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.8759, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.5755, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(70.9723, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.4772, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.8911, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.5583, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(68.9572, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.6246, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.3927, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.1148, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.4685, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(61.0207, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5506, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(108.1316, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(68.3185, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.8315, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.6608, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(96.4335, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.0820, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.8882, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0338, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(144.9031, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(109.4327, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.8134, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.1843, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.8337, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.7099, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.8902, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.6369, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.8816, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.4422, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.0180, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.8452, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.4934, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.5001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.8978, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2825, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.2419, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.8224, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8329, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.4327, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.5898, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.1483, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.6615, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.2638, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.4899, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.4749, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.1389, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.24395604395604395, 'recall': 0.3162393162393162, 'f1': 0.27543424317617865, 'number': 351}, 'P': {'precision': 0.18075801749271136, 'recall': 0.21754385964912282, 'f1': 0.19745222929936307, 'number': 285}, 'overall_precision': 0.21679197994987467, 'overall_recall': 0.2720125786163522, 'overall_f1': 0.2412831241283124, 'overall_accuracy': 0.6749350864593117}
	Train size: 50 Test size: 50


		-------------RUN 1-----------
			------------EPOCH 1---------------
Loss:  tensor(3507.7197, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2229.1018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3602.4092, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3166.8022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1589.2285, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2073.5552, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2131.2856, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2458.3096, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1286.9060, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1528.8422, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2483.9878, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2317.1428, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1175.8938, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2196.1729, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1871.6407, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2365.5557, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1982.8379, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2805.7676, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1881.7957, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3902.7087, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2329.4082, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2461.3076, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1476.7063, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2378.3711, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1282.4248, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1967.1122, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2618.8015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2717.2998, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2462.8809, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2027.2671, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1538.6907, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1258.6731, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2207.2559, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2394.4785, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1869.2483, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.008088978766430738, 'recall': 0.011544011544011544, 'f1': 0.009512485136741973, 'number': 693}, 'P': {'precision': 0.1074983410749834, 'recall': 0.1832579185520362, 'f1': 0.1355081555834379, 'number': 884}, 'overall_precision': 0.06810897435897435, 'overall_recall': 0.10779961953075459, 'overall_f1': 0.08347655290940338, 'overall_accuracy': 0.6100745066884247}
			------------EPOCH 2---------------
Loss:  tensor(2004.1080, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1216.1090, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2274.7759, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1973.3242, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1106.8167, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1410.1311, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1474.6887, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1731.5215, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(970.8940, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1179.9175, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1973.7410, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1834.0935, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(874.2771, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1754.6388, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1697.8386, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1964.1665, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1651.3234, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2370.0295, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1528.5908, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3352.7756, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2034.3461, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2193.0908, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1316.5894, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2164.9121, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1139.3765, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1827.6678, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2313.2876, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2570.4011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2311.3438, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1685.7708, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1279.7065, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1067.1948, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1828.7620, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2222.0581, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1782.5420, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.033268101761252444, 'recall': 0.049062049062049064, 'f1': 0.03965014577259475, 'number': 693}, 'P': {'precision': 0.08773354995938261, 'recall': 0.12217194570135746, 'f1': 0.10212765957446808, 'number': 884}, 'overall_precision': 0.06302707501109632, 'overall_recall': 0.09004438807863031, 'overall_f1': 0.0741514360313316, 'overall_accuracy': 0.625469692740429}
			------------EPOCH 3---------------
Loss:  tensor(1691.4661, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1021.2516, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2009.8607, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1707.6519, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1006.4534, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1221.8660, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1259.5001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1542.6232, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(888.7921, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1062.2358, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1904.4657, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1736.9961, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(739.3392, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1507.0652, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1742.4315, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1884.6476, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1541.1694, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2206.2605, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1239.4734, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3363.6079, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1854.5754, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2081.9971, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1229.2214, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2147.3904, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1022.3345, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1550.6233, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2059.2883, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2165.6978, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2035.3986, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1219.7780, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(851.6648, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(817.9329, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1496.7668, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2213.4375, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1790.7393, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.0673076923076923, 'recall': 0.13131313131313133, 'f1': 0.08899755501222494, 'number': 693}, 'P': {'precision': 0.0440771349862259, 'recall': 0.01809954751131222, 'f1': 0.02566158781074579, 'number': 884}, 'overall_precision': 0.06239067055393586, 'overall_recall': 0.06785034876347495, 'overall_f1': 0.06500607533414339, 'overall_accuracy': 0.5354175165868637}
			------------EPOCH 4---------------
Loss:  tensor(1613.7581, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(987.9265, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1847.4965, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1675.9355, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(943.8943, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1023.5647, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1050.6926, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1455.0249, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(753.6790, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(944.9779, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1599.4580, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1526.4490, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(615.2012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1279.6145, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1480.0332, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1538.2488, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1365.1754, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1934.9658, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1028.9136, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2945.7158, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1570.4866, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1979.7603, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1043.0852, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1932.6904, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(923.0054, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1372.0356, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1900.8069, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1838.5699, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1720.2566, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1046.9827, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(645.7435, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(667.3998, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1200.2493, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1570.7700, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1333.9280, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.10254777070063695, 'recall': 0.23232323232323232, 'f1': 0.14228899690676092, 'number': 693}, 'P': {'precision': 0.043010752688172046, 'recall': 0.03167420814479638, 'f1': 0.03648208469055375, 'number': 884}, 'overall_precision': 0.08509680324178298, 'overall_recall': 0.11984781230183894, 'overall_f1': 0.0995260663507109, 'overall_accuracy': 0.5773087411160973}
			------------EPOCH 5---------------
Loss:  tensor(1207.4065, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(733.1050, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1399.1780, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1375.1963, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(849.6750, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(802.0604, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(853.4660, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1338.7303, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(659.3074, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(854.4789, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1365.7021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1348.1948, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(542.6555, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1176.4583, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1216.8060, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1388.2896, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1130.0938, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1685.4709, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(926.8249, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2420.6689, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1223.8192, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1549.3427, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(842.6707, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1521.0061, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(823.8892, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1175.2939, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1626.2621, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1540.4141, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1496.1262, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(790.9278, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(371.1892, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(491.7278, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(979.9730, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1311.7114, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1087.3989, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.13403880070546736, 'recall': 0.21933621933621933, 'f1': 0.16639299397920088, 'number': 693}, 'P': {'precision': 0.15031645569620253, 'recall': 0.2149321266968326, 'f1': 0.17690875232774672, 'number': 884}, 'overall_precision': 0.14261884904086738, 'overall_recall': 0.21686746987951808, 'overall_f1': 0.1720754716981132, 'overall_accuracy': 0.6552723681102784}
			------------EPOCH 6---------------
Loss:  tensor(923.9062, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(558.5737, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1231.1414, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1287.6255, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(687.8202, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(637.2560, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(663.0768, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1136.6530, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(538.9291, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(709.9964, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1207.9541, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1241.5492, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(421.9062, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(893.4122, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1022.1964, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1006.6667, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(922.9027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1430.8130, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(739.5041, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1896.2307, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(960.8425, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1273.0020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(742.6275, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1182.4050, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(714.2209, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1006.0600, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1377.2775, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1376.5355, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1128.3868, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(531.5613, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(214.0955, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(349.4435, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(698.8481, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1040.1550, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(813.0073, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.16054564533053514, 'recall': 0.22077922077922077, 'f1': 0.1859052247873633, 'number': 693}, 'P': {'precision': 0.1547144754316069, 'recall': 0.26357466063348417, 'f1': 0.19497907949790794, 'number': 884}, 'overall_precision': 0.15697437982919887, 'overall_recall': 0.2447685478757134, 'overall_f1': 0.1912784935579782, 'overall_accuracy': 0.6752839628110708}
			------------EPOCH 7---------------
Loss:  tensor(783.6423, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(418.5621, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(859.7025, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1011.8873, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(638.7498, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(504.4753, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(495.6615, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1014.1588, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(547.3877, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(656.1779, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1270.0883, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1207.6436, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(316.7357, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(659.5612, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1113.6633, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(912.4609, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(630.1653, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1015.0015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(580.7247, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1520.2825, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(712.3005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(948.3960, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(533.1749, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(928.4237, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(621.2508, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(928.5718, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1242.7048, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1298.9070, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1283.6536, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(443.5981, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(201.4050, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(353.9222, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(767.1943, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1182.2197, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(973.8887, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.15837563451776648, 'recall': 0.22510822510822512, 'f1': 0.18593563766388554, 'number': 693}, 'P': {'precision': 0.09568345323741007, 'recall': 0.1504524886877828, 'f1': 0.1169744942832014, 'number': 884}, 'overall_precision': 0.12168421052631578, 'overall_recall': 0.18325935320228282, 'overall_f1': 0.14625506072874495, 'overall_accuracy': 0.67627165954523}
			------------EPOCH 8---------------
Loss:  tensor(575.7827, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(290.8095, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(679.8629, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(750.0192, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(475.8124, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(344.0065, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(298.8892, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(677.1863, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(413.5552, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(507.9684, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(935.5120, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1076.0804, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(275.8066, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(563.3752, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1094.8267, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(850.2520, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(596.1218, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1004.6163, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(653.3318, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1930.4460, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(762.1016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1278.2065, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(888.8910, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1630.0679, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(599.5496, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(773.4263, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1173.7870, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(997.9703, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(883.3541, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(298.9591, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(144.5258, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(250.7831, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(480.4735, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(794.7361, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(666.2225, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.11981981981981982, 'recall': 0.1919191919191919, 'f1': 0.1475318912922906, 'number': 693}, 'P': {'precision': 0.039297658862876256, 'recall': 0.053167420814479636, 'f1': 0.04519230769230769, 'number': 884}, 'overall_precision': 0.07805724197745013, 'overall_recall': 0.11414077362079898, 'overall_f1': 0.09271182075714653, 'overall_accuracy': 0.6097739033345501}
			------------EPOCH 9---------------
Loss:  tensor(590.9482, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(296.3217, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(678.4634, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(704.3197, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(475.3989, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(442.9552, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(371.9087, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(832.3356, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(448.9446, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(573.1011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1253.8799, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1120.5033, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(305.3193, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(525.2351, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(897.5959, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(759.1011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(477.2844, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(781.8064, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(390.5942, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1134.6287, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(507.8121, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(775.2424, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(401.7885, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(740.0091, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(507.6288, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(649.1080, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(943.1776, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(747.8577, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(721.8363, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(312.0610, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(131.4631, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(215.0415, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(597.2112, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(838.7279, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(642.6650, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.15862944162436549, 'recall': 0.36075036075036077, 'f1': 0.22036139268400176, 'number': 693}, 'P': {'precision': 0.17420212765957446, 'recall': 0.14819004524886878, 'f1': 0.16014669926650368, 'number': 884}, 'overall_precision': 0.163659793814433, 'overall_recall': 0.24159797083069118, 'overall_f1': 0.19513444302176697, 'overall_accuracy': 0.6146265003328109}
			------------EPOCH 10---------------
Loss:  tensor(543.8804, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(282.6176, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(707.1592, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(941.0056, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(477.0304, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(273.7274, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(276.6131, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(735.0823, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(348.3064, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(321.3662, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(666.4581, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(760.7805, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(192.5850, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(298.3591, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(601.6609, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(386.6526, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(340.4377, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(626.7018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(303.8997, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1137.5039, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(510.8471, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(794.9796, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(401.2832, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(643.3719, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(400.5007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(534.3269, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(829.5917, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(680.0213, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(485.4781, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(214.4012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(98.5780, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(183.7198, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(349.2599, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(571.6278, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(494.7599, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.1252096143096702, 'recall': 0.32323232323232326, 'f1': 0.18049959709911362, 'number': 693}, 'P': {'precision': 0.03205128205128205, 'recall': 0.016968325791855202, 'f1': 0.022189349112426034, 'number': 884}, 'overall_precision': 0.10589277802392556, 'overall_recall': 0.15155358275206088, 'overall_f1': 0.1246739697443923, 'overall_accuracy': 0.5731217658299873}
			------------EPOCH 11---------------
Loss:  tensor(421.7869, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(207.0582, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(460.9001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(590.6607, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(469.6675, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(217.0959, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(264.9932, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(626.9111, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(287.6884, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(324.3536, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(629.9086, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(874.6600, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(168.0544, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(296.7085, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(432.1290, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(553.5280, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(281.4617, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(567.4612, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(250.1631, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(972.7423, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(351.9677, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(509.5372, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(265.2828, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(435.5373, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(399.3807, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(598.4345, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(771.9030, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(604.5660, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(700.2450, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(417.6359, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(140.9605, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(216.0408, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(493.5110, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(871.7100, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(546.3632, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.1743827160493827, 'recall': 0.16305916305916307, 'f1': 0.168530947054437, 'number': 693}, 'P': {'precision': 0.13159355913381454, 'recall': 0.2680995475113122, 'f1': 0.17653631284916202, 'number': 884}, 'overall_precision': 0.14291547570436913, 'overall_recall': 0.2219403931515536, 'overall_f1': 0.17386984600099356, 'overall_accuracy': 0.6926974856676615}
			------------EPOCH 12---------------
Loss:  tensor(384.0369, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(149.4688, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(425.3878, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(503.6080, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(218.5500, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(149.3188, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(153.0500, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(294.4738, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(186.8166, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(198.3764, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(361.2466, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(551.4181, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(147.0748, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(231.7189, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(337.2089, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(292.1384, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(323.0767, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(590.0064, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(355.5092, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(948.7689, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(347.4992, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(631.5511, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(362.0632, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(599.5367, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(330.6560, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(560.7776, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1099.8013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(751.6282, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(409.6862, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(150.4599, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(72.6828, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(123.8385, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(256.2342, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(387.9771, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(263.6826, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.18995929443690637, 'recall': 0.20202020202020202, 'f1': 0.19580419580419578, 'number': 693}, 'P': {'precision': 0.10773993808049535, 'recall': 0.19683257918552036, 'f1': 0.13925570228091236, 'number': 884}, 'overall_precision': 0.13350340136054423, 'overall_recall': 0.19911223842739378, 'overall_f1': 0.1598371086790532, 'overall_accuracy': 0.6952096708393275}
			------------EPOCH 13---------------
Loss:  tensor(224.7213, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(97.0050, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(305.7884, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(310.4496, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(203.3915, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(166.8961, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(160.8793, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(294.8709, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(230.9470, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(249.0234, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(580.9368, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(874.5012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(139.9569, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(267.5680, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(467.3676, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(395.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(234.2644, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(537.2410, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(181.7699, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(929.6055, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(232.7730, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(352.7820, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(219.3352, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(335.2778, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(199.3590, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(310.5821, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(555.6577, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(348.1964, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(255.4691, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(123.0771, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(65.0380, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(97.7921, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(205.4026, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(332.5656, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(227.1873, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.14747736093143596, 'recall': 0.329004329004329, 'f1': 0.20366234926306384, 'number': 693}, 'P': {'precision': 0.11838006230529595, 'recall': 0.12895927601809956, 'f1': 0.12344342176502437, 'number': 884}, 'overall_precision': 0.13630928656835392, 'overall_recall': 0.21686746987951808, 'overall_f1': 0.16740088105726872, 'overall_accuracy': 0.65495029308827}
			------------EPOCH 14---------------
Loss:  tensor(166.5418, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(77.4697, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(236.6175, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(298.2055, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(206.6021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(78.8434, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(74.8000, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(236.6183, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(175.3018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(173.5620, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(272.1416, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(441.1354, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(121.7049, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(176.3847, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(249.0404, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(223.9008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(149.3027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(319.5477, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(169.8885, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(479.4404, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(173.6319, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(285.1338, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(169.3563, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(252.4865, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(170.3911, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(250.3658, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(447.0546, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(284.3162, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(208.7545, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(106.9787, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.7531, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(80.8736, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(148.9340, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(233.0640, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(129.9330, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.17406542056074767, 'recall': 0.215007215007215, 'f1': 0.19238218205293736, 'number': 693}, 'P': {'precision': 0.13522884882108183, 'recall': 0.22058823529411764, 'f1': 0.16766981943250214, 'number': 884}, 'overall_precision': 0.1496953872932985, 'overall_recall': 0.21813570069752694, 'overall_f1': 0.17754838709677417, 'overall_accuracy': 0.6895196787838447}
			------------EPOCH 15---------------
Loss:  tensor(113.4325, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.2871, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(158.4136, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(197.6186, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(112.7153, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.1751, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(65.1577, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(135.9399, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(96.1709, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(120.7268, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(191.5608, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(368.3281, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(62.1493, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(104.2438, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(234.1803, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(154.9944, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(113.9325, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(214.2258, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(77.6517, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(447.0743, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(138.5617, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(239.1858, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(144.6071, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(191.1092, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(124.4660, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(181.4513, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(368.9603, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(208.1487, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(146.9989, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(79.7385, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.5844, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.6877, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(120.6850, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(178.6552, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(91.2787, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.1696344892221181, 'recall': 0.2611832611832612, 'f1': 0.2056818181818182, 'number': 693}, 'P': {'precision': 0.1292372881355932, 'recall': 0.20701357466063347, 'f1': 0.1591304347826087, 'number': 884}, 'overall_precision': 0.14659685863874344, 'overall_recall': 0.23081800887761572, 'overall_f1': 0.1793103448275862, 'overall_accuracy': 0.6862774568956262}
			------------EPOCH 16---------------
Loss:  tensor(90.2928, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.0612, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(131.1190, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(149.9899, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(92.7217, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.8367, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.0602, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(87.7103, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(77.3438, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(86.0398, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(128.3205, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(250.1003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.3147, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(71.9643, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(160.4542, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(125.3621, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(89.5976, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(190.2752, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(66.8858, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(280.3396, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(77.3213, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(165.9411, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(104.6334, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(148.5840, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(85.7625, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(141.6603, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(300.2977, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(159.3335, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(122.1915, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.7051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.8769, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.7790, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(87.6420, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(134.6483, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(69.4517, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.1811248808388942, 'recall': 0.2741702741702742, 'f1': 0.21814006888633755, 'number': 693}, 'P': {'precision': 0.14473684210526316, 'recall': 0.19909502262443438, 'f1': 0.1676190476190476, 'number': 884}, 'overall_precision': 0.16158940397350993, 'overall_recall': 0.2320862396956246, 'overall_f1': 0.19052576782925562, 'overall_accuracy': 0.6823051982908552}
			------------EPOCH 17---------------
Loss:  tensor(66.1695, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.3701, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(109.3465, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(109.7241, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.8544, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.2326, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.6178, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(61.6969, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.5385, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(66.4278, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(95.1690, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(190.6368, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.0737, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.7495, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(140.6748, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(80.1239, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.8224, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(130.6331, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.0894, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(238.1494, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(67.5698, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(139.7824, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(85.7347, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(107.1969, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(61.0175, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(102.6785, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(265.1436, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(135.0548, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(99.7024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.5230, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.5030, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.2976, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(67.7812, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(107.4698, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.9453, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.16666666666666666, 'recall': 0.2481962481962482, 'f1': 0.19942028985507246, 'number': 693}, 'P': {'precision': 0.13049853372434017, 'recall': 0.20135746606334842, 'f1': 0.1583629893238434, 'number': 884}, 'overall_precision': 0.14607679465776294, 'overall_recall': 0.2219403931515536, 'overall_f1': 0.17618927762396172, 'overall_accuracy': 0.6869001352715093}
			------------EPOCH 18---------------
Loss:  tensor(50.0962, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.1931, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(80.9413, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(84.5383, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.0285, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.0094, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.3593, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.1007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.2132, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.0828, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(69.6522, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(139.8993, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.3485, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.8817, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(107.7338, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(59.6338, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.1784, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(110.1804, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.2526, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(206.3379, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.9580, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(101.9401, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(69.5338, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(86.4650, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.0255, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(80.9016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(224.1518, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(107.2343, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(70.0458, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.5699, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.1585, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.1666, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(59.7800, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(97.3180, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.3200, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.19702970297029704, 'recall': 0.28715728715728717, 'f1': 0.2337052260716383, 'number': 693}, 'P': {'precision': 0.15289575289575288, 'recall': 0.2239819004524887, 'f1': 0.1817347407067462, 'number': 884}, 'overall_precision': 0.172234273318872, 'overall_recall': 0.2517438173747622, 'overall_f1': 0.2045337454920144, 'overall_accuracy': 0.687995190346338}
			------------EPOCH 19---------------
Loss:  tensor(40.3270, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.5337, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(74.9725, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(72.4592, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.1468, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.2637, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.3460, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.8646, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.9637, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.5103, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.8895, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(110.3197, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.9531, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.0570, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(90.4733, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.2607, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.7706, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(82.1839, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.1360, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(164.0522, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.3589, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(78.1969, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.7496, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(65.5786, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.0376, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.8109, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(224.7928, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(86.9166, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.5448, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.1257, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.1949, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.3132, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.4760, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(86.9199, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.2975, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.16602316602316602, 'recall': 0.2481962481962482, 'f1': 0.19895893580104107, 'number': 693}, 'P': {'precision': 0.13975155279503104, 'recall': 0.20361990950226244, 'f1': 0.16574585635359115, 'number': 884}, 'overall_precision': 0.1514629948364888, 'overall_recall': 0.22320862396956245, 'overall_f1': 0.1804665470392207, 'overall_accuracy': 0.6826487449809976}
			------------EPOCH 20---------------
Loss:  tensor(31.7365, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.6668, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(86.4519, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.6755, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.0565, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.6382, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.5990, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.7202, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.3072, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.5406, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.6918, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(89.3746, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.3466, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.9264, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(69.7770, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.2479, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.5833, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(74.0250, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.7578, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(131.0486, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.3563, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.0379, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.7343, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(69.6422, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.3332, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.9789, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(172.5740, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(81.3899, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(67.5271, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.6063, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.4780, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.6086, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(78.3764, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(145.8817, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.9380, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.16734279918864098, 'recall': 0.23809523809523808, 'f1': 0.19654556283502084, 'number': 693}, 'P': {'precision': 0.12850812407680945, 'recall': 0.19683257918552036, 'f1': 0.15549597855227884, 'number': 884}, 'overall_precision': 0.14487179487179488, 'overall_recall': 0.21496512365250475, 'overall_f1': 0.17309165177431707, 'overall_accuracy': 0.68591243853735}
			------------EPOCH 21---------------
Loss:  tensor(24.4557, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.4248, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(96.2040, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.9236, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.2814, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.4561, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.3391, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.1284, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.1516, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.3042, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.2332, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(82.3725, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.2036, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.2102, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.0418, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.6663, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.0931, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.5120, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.6636, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(188.1493, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.8151, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.2491, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.7297, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(64.9102, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.5281, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.5411, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(163.9148, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(76.4998, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(65.3125, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.6540, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.7791, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.6529, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(62.9642, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(116.2899, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.8757, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.18619436875567666, 'recall': 0.2958152958152958, 'f1': 0.22853957636566333, 'number': 693}, 'P': {'precision': 0.15847860538827258, 'recall': 0.22624434389140272, 'f1': 0.1863932898415657, 'number': 884}, 'overall_precision': 0.1713922979263648, 'overall_recall': 0.2568167406467977, 'overall_f1': 0.20558375634517764, 'overall_accuracy': 0.6832928950250144}
			------------EPOCH 22---------------
Loss:  tensor(27.7701, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.3645, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(108.8293, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.7448, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.7790, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.6470, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.5855, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.8525, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.5127, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.4771, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.6187, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(96.5876, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.6303, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.1940, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.2783, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(62.3868, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.8393, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(89.6298, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.2770, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(174.8547, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.7898, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.9842, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(69.3443, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(80.9004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.2490, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.6089, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(142.2625, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(92.8496, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(88.9606, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.5006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.5490, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.3797, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.8822, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(88.9992, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.5878, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.17027863777089783, 'recall': 0.23809523809523808, 'f1': 0.19855595667870035, 'number': 693}, 'P': {'precision': 0.1288732394366197, 'recall': 0.20701357466063347, 'f1': 0.15885416666666666, 'number': 884}, 'overall_precision': 0.14566764336542487, 'overall_recall': 0.2206721623335447, 'overall_f1': 0.17549167927382753, 'overall_accuracy': 0.6850106284757262}
			------------EPOCH 23---------------
Loss:  tensor(26.0162, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.3673, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(86.7359, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(64.0566, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.0858, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.1330, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.7619, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.2446, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.0785, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.3714, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.2871, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(59.4195, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.6116, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.0238, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.2382, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.8169, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.6944, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(68.7685, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.8296, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(149.6799, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.3409, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.9759, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.2212, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.6010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.6071, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.9370, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(164.6947, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(82.5217, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.8000, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.1855, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.9872, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.0626, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.6307, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.8933, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.2167, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.1781781781781782, 'recall': 0.25685425685425683, 'f1': 0.21040189125295508, 'number': 693}, 'P': {'precision': 0.14650145772594753, 'recall': 0.22737556561085973, 'f1': 0.17819148936170212, 'number': 884}, 'overall_precision': 0.15984816533108392, 'overall_recall': 0.2403297400126823, 'overall_f1': 0.19199594731509623, 'overall_accuracy': 0.6907865072037447}
			------------EPOCH 24---------------
Loss:  tensor(21.3409, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.4911, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.2153, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.3264, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.1209, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.7109, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.0633, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.1842, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.4980, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.8830, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.0477, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.4659, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.3348, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.2815, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.9074, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.1466, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.8955, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.7454, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.8743, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(125.2132, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.6113, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.7114, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.2165, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(59.3927, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.2389, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.4427, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(107.3494, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(79.0503, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.3598, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.8902, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.5493, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.1795, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.1035, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.6767, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.1031, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.188050930460333, 'recall': 0.27705627705627706, 'f1': 0.22403733955659277, 'number': 693}, 'P': {'precision': 0.15710919088766692, 'recall': 0.22624434389140272, 'f1': 0.18544274455261936, 'number': 884}, 'overall_precision': 0.17088055797733218, 'overall_recall': 0.24857324032974001, 'overall_f1': 0.20253164556962028, 'overall_accuracy': 0.6843020634273077}
			------------EPOCH 25---------------
Loss:  tensor(19.6417, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.0850, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.3602, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.2291, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.4236, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.5397, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.9407, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.2594, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.3715, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.8774, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.7162, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.4502, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.8717, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.9197, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.0040, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.4177, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.9270, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.3700, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.2621, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(109.4818, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.1048, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.0489, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.5288, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(69.1438, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.9805, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.0709, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(88.4498, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.3085, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.6517, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.8102, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.3253, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.8858, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.8869, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.1331, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.5392, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.18048268625393493, 'recall': 0.2481962481962482, 'f1': 0.20899149453219926, 'number': 693}, 'P': {'precision': 0.14200298953662183, 'recall': 0.2149321266968326, 'f1': 0.17101710171017104, 'number': 884}, 'overall_precision': 0.15800960279353993, 'overall_recall': 0.22954977805960686, 'overall_f1': 0.18717683557394002, 'overall_accuracy': 0.687995190346338}
			------------EPOCH 26---------------
Loss:  tensor(9.1629, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.0733, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.1731, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.2117, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.3499, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.2582, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.1319, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.8169, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.6192, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.3055, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.1255, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.9937, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.0556, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.2039, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.4099, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.9123, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.1199, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.3950, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.0182, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(96.3452, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.6296, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.9363, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.0352, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.4069, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.9584, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.2578, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(65.7932, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.3371, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.9128, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.9473, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.7834, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.7915, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.3427, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.7451, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.9785, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.18357933579335795, 'recall': 0.28715728715728717, 'f1': 0.2239729881823298, 'number': 693}, 'P': {'precision': 0.15926493108728942, 'recall': 0.23529411764705882, 'f1': 0.18995433789954336, 'number': 884}, 'overall_precision': 0.17029288702928871, 'overall_recall': 0.2580849714648066, 'overall_f1': 0.20519284093773635, 'overall_accuracy': 0.6870719086165804}
			------------EPOCH 27---------------
Loss:  tensor(9.1198, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2660, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.8359, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.8511, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.4897, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.3465, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.0974, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.6529, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.3271, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.9505, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.6382, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.5220, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.2634, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.2640, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.6300, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.9781, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.4694, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.6214, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.5769, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(86.0823, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.6547, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.2779, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.3107, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.2744, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.3199, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.9916, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(67.6120, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.8004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.1935, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.5171, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.0471, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.7137, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.2942, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.5191, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.7100, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.17938144329896907, 'recall': 0.2510822510822511, 'f1': 0.20926037282020446, 'number': 693}, 'P': {'precision': 0.14690529455630127, 'recall': 0.2228506787330317, 'f1': 0.17707865168539327, 'number': 884}, 'overall_precision': 0.160536564257897, 'overall_recall': 0.2352568167406468, 'overall_f1': 0.19084362139917696, 'overall_accuracy': 0.6899920554827904}
			------------EPOCH 28---------------
Loss:  tensor(5.8648, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.0854, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.3317, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.2300, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.8087, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.8349, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.9126, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.0948, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.5669, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.0993, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.7522, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.9382, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.8095, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.2705, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.6960, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.1219, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.2853, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.8785, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.8319, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(59.0165, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.8405, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.4388, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.2503, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.4151, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.7334, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.7109, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.4315, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.5995, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.0463, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.1129, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.4750, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.2438, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.9949, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.0476, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.0883, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.1819137749737119, 'recall': 0.24963924963924963, 'f1': 0.2104622871046229, 'number': 693}, 'P': {'precision': 0.1395007342143906, 'recall': 0.2149321266968326, 'f1': 0.16918967052537848, 'number': 884}, 'overall_precision': 0.1569390402075227, 'overall_recall': 0.2301838934686113, 'overall_f1': 0.18663239074550128, 'overall_accuracy': 0.6919459772829751}
			------------EPOCH 29---------------
Loss:  tensor(4.8564, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.8413, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.4733, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.5589, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.4598, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.3530, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.4602, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.5809, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.7746, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.5964, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.4005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.2289, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.3260, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.9796, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.0140, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.4528, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.2552, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.1741, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.8743, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.7723, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.2840, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.7691, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.4856, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.0133, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.1643, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.5154, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.7843, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.8865, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.6333, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.7459, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.3821, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.4528, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.2365, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.5270, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.1518, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.18635437881873726, 'recall': 0.26406926406926406, 'f1': 0.21850746268656718, 'number': 693}, 'P': {'precision': 0.13929618768328444, 'recall': 0.2149321266968326, 'f1': 0.16903914590747332, 'number': 884}, 'overall_precision': 0.15899403239556692, 'overall_recall': 0.23652504755865567, 'overall_f1': 0.1901605913841448, 'overall_accuracy': 0.6907650355356107}
			------------EPOCH 30---------------
Loss:  tensor(4.1993, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5674, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.1821, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.8331, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.1066, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.1682, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.1172, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.9324, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.3958, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.3094, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.5920, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.6821, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.0431, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.6880, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.5746, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.6762, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.7858, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.6969, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.6806, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.1272, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.4607, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.3886, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.2499, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.4664, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.3076, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.1630, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.8180, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.7315, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.4722, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.0863, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.9017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.9494, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.0830, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.7167, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.6321, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.17813765182186234, 'recall': 0.25396825396825395, 'f1': 0.2093991671624033, 'number': 693}, 'P': {'precision': 0.14147431124348472, 'recall': 0.2149321266968326, 'f1': 0.17063313875168387, 'number': 884}, 'overall_precision': 0.15701415701415702, 'overall_recall': 0.2320862396956246, 'overall_f1': 0.1873080859774821, 'overall_accuracy': 0.6890258304167651}


		-------------RUN 2-----------
			------------EPOCH 1---------------
Loss:  tensor(2786.8726, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2205.6968, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2939.7437, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1790.1145, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2767.9045, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1659.2344, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2692.2659, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1663.4869, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2392.1917, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2969.3274, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2749.7048, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3495.3472, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1766.1801, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2594.7534, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1509.8379, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2587.9863, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2161.6292, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1132.5101, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1413.1624, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1364.9924, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2865.9883, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2858.4785, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1983.4430, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1836.1736, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2127.9375, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2348.4717, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2458.0164, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2231.8545, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2395.3506, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(990.9316, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(819.6775, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1256.0491, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1037.3188, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(943.8425, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1053.7454, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1911.4583, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.0234375, 'recall': 0.02763157894736842, 'f1': 0.02536231884057971, 'number': 760}, 'P': {'precision': 0.06256742179072276, 'recall': 0.07125307125307126, 'f1': 0.0666283744974153, 'number': 814}, 'overall_precision': 0.04333516182117389, 'overall_recall': 0.05019059720457433, 'overall_f1': 0.046511627906976744, 'overall_accuracy': 0.5607673241548812}
			------------EPOCH 2---------------
Loss:  tensor(1650.2539, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1296.2244, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1839.6768, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1119.0978, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1796.8494, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1090.7097, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1704.7894, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1132.1194, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1774.5776, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2324.5684, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2033.6162, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2489.8682, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1207.2383, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2045.2404, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1007.4914, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1766.3335, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2003.4247, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(861.7660, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1135.3623, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1063.6326, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2321.6211, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2298.4590, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1638.4603, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1360.9697, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1861.5444, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1877.0540, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1806.0812, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2028.1284, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2106.0198, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(863.3198, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(660.5222, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(984.6451, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(800.8171, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(790.3162, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(931.1329, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1649.5646, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.04243119266055046, 'recall': 0.04868421052631579, 'f1': 0.045343137254901966, 'number': 760}, 'P': {'precision': 0.1157613535173642, 'recall': 0.1597051597051597, 'f1': 0.1342281879194631, 'number': 814}, 'overall_precision': 0.0837092731829574, 'overall_recall': 0.10609911054637865, 'overall_f1': 0.0935836368730737, 'overall_accuracy': 0.6057153568973777}
			------------EPOCH 3---------------
Loss:  tensor(1440.2556, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1129.4751, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1584.5588, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(877.1290, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1473.9958, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(890.5193, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1415.4509, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(970.0109, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1380.9243, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1911.5632, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1720.6350, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2099.1958, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(944.1660, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1620.9170, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(854.6027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1526.6240, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1776.5115, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(705.7389, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(944.4362, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(893.1760, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1947.0911, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1906.9519, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1410.7175, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1123.8766, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1564.2198, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1618.1775, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1381.4531, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1808.2271, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1748.6899, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(717.4196, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(572.5793, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(762.7262, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(617.2198, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(627.9493, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(821.0198, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1335.1187, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.07790368271954674, 'recall': 0.07236842105263158, 'f1': 0.07503410641200546, 'number': 760}, 'P': {'precision': 0.1891025641025641, 'recall': 0.28992628992628994, 'f1': 0.2289039767216295, 'number': 814}, 'overall_precision': 0.1489252814738997, 'overall_recall': 0.1848792884371029, 'overall_f1': 0.1649659863945578, 'overall_accuracy': 0.6193997208972943}
			------------EPOCH 4---------------
Loss:  tensor(1196.0159, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(944.7114, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1386.5391, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(752.7363, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1212.4879, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(656.0258, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1161.1643, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(782.3955, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(983.8950, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1428.4558, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1322.2361, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1715.8806, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(700.5836, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1229.0891, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(743.7090, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1263.2928, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1357.6582, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(578.0018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(737.9661, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(736.4243, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1615.3827, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1633.9773, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1159.2067, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(924.2270, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1216.2562, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1334.1744, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1002.6852, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1490.8265, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1386.5804, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(556.1074, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(478.5686, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(563.9686, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(448.9073, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(476.9905, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(644.1236, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1170.3871, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.14014752370916755, 'recall': 0.175, 'f1': 0.15564657694558223, 'number': 760}, 'P': {'precision': 0.15996578272027373, 'recall': 0.22972972972972974, 'f1': 0.18860312657589512, 'number': 814}, 'overall_precision': 0.1510859301227573, 'overall_recall': 0.20330368487928843, 'overall_f1': 0.1733477789815818, 'overall_accuracy': 0.6349794838682802}
			------------EPOCH 5---------------
Loss:  tensor(1095.1118, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(866.2122, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1157.8162, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(554.0526, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(945.4596, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(536.2862, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(963.2420, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(701.5209, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(820.2275, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1273.5737, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1206.0895, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1633.4131, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(526.9020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(934.4544, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(649.3444, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1068.4604, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1122.7522, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(493.0412, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(633.5837, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(618.1437, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1767.0847, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1829.4453, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1299.3416, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1047.3098, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1232.3494, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1228.4468, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(931.6364, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1421.4023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1289.6057, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(452.1416, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(442.2659, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(413.3383, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(374.8066, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(366.0174, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(553.5034, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1010.0105, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.2196969696969697, 'recall': 0.11447368421052631, 'f1': 0.15051903114186851, 'number': 760}, 'P': {'precision': 0.19129878438899553, 'recall': 0.36732186732186733, 'f1': 0.2515776188472865, 'number': 814}, 'overall_precision': 0.1970393057682491, 'overall_recall': 0.24523506988564167, 'overall_f1': 0.21851118030002828, 'overall_accuracy': 0.6023619587177939}
			------------EPOCH 6---------------
Loss:  tensor(1407.8457, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(984.2507, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1489.1443, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(595.6055, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1048.6255, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(545.9216, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(988.6111, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(756.7869, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(793.4678, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1173.3799, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1074.6945, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1653.8877, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(493.6228, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(880.9135, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(655.7993, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1018.2307, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1169.7869, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(428.2731, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(599.4374, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(526.9262, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1304.8677, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1467.1724, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(985.4863, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(699.3102, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1105.7512, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1158.5701, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1063.6460, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1501.4514, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1300.2930, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(591.0262, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(499.7651, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(623.3835, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(418.5955, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(502.1184, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(586.3557, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1478.8496, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.15487094088259784, 'recall': 0.24473684210526317, 'f1': 0.1896991330953595, 'number': 760}, 'P': {'precision': 0.08075370121130551, 'recall': 0.07371007371007371, 'f1': 0.07707129094412331, 'number': 814}, 'overall_precision': 0.12654320987654322, 'overall_recall': 0.156289707750953, 'overall_f1': 0.13985218874360436, 'overall_accuracy': 0.6215034054695799}
			------------EPOCH 7---------------
Loss:  tensor(745.1476, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(734.3369, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1114.6150, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(756.6769, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(862.4434, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(431.1321, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(862.3677, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(650.3081, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(553.9082, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(845.2542, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(793.2533, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1202.9585, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(361.2000, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(752.7916, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(560.2419, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(842.8923, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1290.8879, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(517.9301, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(760.6796, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(592.0743, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1853.8207, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1625.1936, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1261.6614, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1069.4552, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1134.1814, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1221.8491, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(830.7247, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1373.7075, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1029.6157, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(353.6745, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(362.9865, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(316.8561, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(209.1681, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(272.5978, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(381.5244, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(829.9579, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.12825520833333334, 'recall': 0.25921052631578945, 'f1': 0.171602787456446, 'number': 760}, 'P': {'precision': 0.0810126582278481, 'recall': 0.03931203931203931, 'f1': 0.05293631100082713, 'number': 814}, 'overall_precision': 0.11859140341791817, 'overall_recall': 0.1454891994917408, 'overall_f1': 0.13067047075606278, 'overall_accuracy': 0.5969673616462894}
			------------EPOCH 8---------------
Loss:  tensor(624.4012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(643.1428, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(931.3969, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(539.8811, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1047.9999, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(456.2186, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(881.4865, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(726.7734, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(780.7343, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1303.3884, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1048.8796, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1718.2588, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(509.7029, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1003.6898, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(682.3782, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1124.5154, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(827.6301, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(301.0792, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(421.5096, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(369.8809, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1079.6989, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1099.9244, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(723.9219, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(495.3540, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(752.6711, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(804.9985, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(606.5939, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(940.6153, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(868.2791, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(309.3798, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(351.4332, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(308.1474, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(216.2900, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(238.3256, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(432.9996, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(875.3696, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.2344139650872818, 'recall': 0.12368421052631579, 'f1': 0.1619293712316968, 'number': 760}, 'P': {'precision': 0.15928066795118817, 'recall': 0.3046683046683047, 'f1': 0.20919443272880642, 'number': 814}, 'overall_precision': 0.17466802860061287, 'overall_recall': 0.21728081321473952, 'overall_f1': 0.19365798414496035, 'overall_accuracy': 0.6524963029305784}
			------------EPOCH 9---------------
Loss:  tensor(727.8645, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(551.0774, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(874.7747, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(445.8542, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(661.8150, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(334.2781, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(604.1260, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(433.8571, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(406.6445, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(646.5873, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(742.7001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1048.3896, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(281.4612, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(575.6665, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(472.9793, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(681.3691, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(702.3316, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(236.7842, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(354.1492, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(320.0213, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(876.1172, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1011.8009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(596.3687, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(393.9215, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(574.3291, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(562.2606, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(508.7227, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(855.4396, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(735.3017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(249.4422, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(264.1640, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(202.2307, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(149.6788, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(236.4164, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(328.7799, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(719.0911, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.15089722675367048, 'recall': 0.24342105263157895, 'f1': 0.1863041289023162, 'number': 760}, 'P': {'precision': 0.07127882599580712, 'recall': 0.08353808353808354, 'f1': 0.07692307692307691, 'number': 814}, 'overall_precision': 0.11605504587155964, 'overall_recall': 0.16073697585768743, 'overall_f1': 0.134789557805008, 'overall_accuracy': 0.6409572806231906}
			------------EPOCH 10---------------
Loss:  tensor(418.7653, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(497.2902, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(578.8602, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(310.3842, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(526.2014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(206.4718, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(436.2889, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(399.5588, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(333.5479, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(589.2427, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(446.6271, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(798.1229, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(194.6714, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(397.6904, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(346.3969, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(460.4811, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(508.4215, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(172.8365, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(257.0309, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(250.1244, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(730.4807, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(828.2015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(459.0359, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(304.7480, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(460.0863, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(486.0604, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(366.1851, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(632.0760, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(593.6075, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(189.2576, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(254.8792, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(159.3286, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(106.8553, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(132.3297, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(246.1186, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(510.4205, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.17029972752043596, 'recall': 0.16447368421052633, 'f1': 0.16733601070950466, 'number': 760}, 'P': {'precision': 0.12232204561161023, 'recall': 0.21744471744471744, 'f1': 0.15656789031402032, 'number': 814}, 'overall_precision': 0.13846859238881248, 'overall_recall': 0.19186785260482847, 'overall_f1': 0.16085219707057258, 'overall_accuracy': 0.6728874632896628}
			------------EPOCH 11---------------
Loss:  tensor(341.7050, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(296.2767, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(432.6049, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(207.1173, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(411.2815, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(131.4039, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(301.7333, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(215.2074, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(205.6239, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(373.4717, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(335.6636, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(567.0734, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(113.9100, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(292.1299, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(250.0707, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(338.0515, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(399.9583, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(133.4537, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(188.3839, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(200.4295, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(592.4944, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(702.1523, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(364.5188, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(240.0665, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(365.8472, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(366.3581, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(373.6530, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(546.2015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(449.4478, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(146.5544, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(176.7577, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(105.9640, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.8471, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(96.9926, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(195.6874, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(353.0141, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.18144611186903137, 'recall': 0.175, 'f1': 0.1781647689216343, 'number': 760}, 'P': {'precision': 0.10411140583554376, 'recall': 0.1928746928746929, 'f1': 0.1352282515073213, 'number': 814}, 'overall_precision': 0.12940651494868363, 'overall_recall': 0.18424396442185514, 'overall_f1': 0.15203145478374835, 'overall_accuracy': 0.6787402886838433}
			------------EPOCH 12---------------
Loss:  tensor(234.8468, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(217.2026, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(298.2369, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(146.9516, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(343.0833, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(83.0898, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(247.9249, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(137.7646, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(148.6688, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(294.9611, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(213.1801, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(433.4315, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(74.0508, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(230.7472, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(197.6721, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(271.4691, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(308.4602, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(100.4073, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(146.8336, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(153.2173, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(505.5911, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(574.9236, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(306.9514, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(180.5429, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(296.3037, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(269.1007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(325.0617, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(417.0548, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(366.3327, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(124.2676, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(151.8134, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(88.8557, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.1032, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(89.8589, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(179.5487, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(312.9980, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.1651676206050695, 'recall': 0.2657894736842105, 'f1': 0.2037317196167423, 'number': 760}, 'P': {'precision': 0.1261930010604454, 'recall': 0.14619164619164618, 'f1': 0.1354581673306773, 'number': 814}, 'overall_precision': 0.1481994459833795, 'overall_recall': 0.2039390088945362, 'overall_f1': 0.17165775401069516, 'overall_accuracy': 0.6553914727874862}
			------------EPOCH 13---------------
Loss:  tensor(180.3160, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(206.0542, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(272.2971, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(147.9866, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(271.7043, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(72.1171, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(208.1117, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(120.0235, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(138.0163, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(238.5359, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(153.7832, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(381.8908, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(66.5823, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(221.1619, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(169.4061, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(258.6224, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(259.3997, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(99.3613, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(126.3327, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(141.9226, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(371.1852, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(439.5444, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(226.1064, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(120.0556, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(223.4744, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(210.7110, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(273.4263, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(344.9473, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(315.3693, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(88.8723, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(155.8793, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(77.9051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.2140, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(75.0438, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(174.4139, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(280.7208, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.20118343195266272, 'recall': 0.2236842105263158, 'f1': 0.2118380062305296, 'number': 760}, 'P': {'precision': 0.1781437125748503, 'recall': 0.29238329238329236, 'f1': 0.22139534883720927, 'number': 814}, 'overall_precision': 0.18707015130674004, 'overall_recall': 0.25921219822109276, 'overall_f1': 0.21731025299600532, 'overall_accuracy': 0.665555810126846}
			------------EPOCH 14---------------
Loss:  tensor(182.8145, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(166.1486, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(265.9745, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(139.5193, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(253.3930, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(65.1322, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(162.1529, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(155.5963, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(115.5424, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(180.2123, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(116.7828, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(263.2598, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.6129, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(180.3196, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(136.7863, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(198.3719, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(199.9945, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(85.4252, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(94.6556, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(100.1112, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(380.1911, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(431.1207, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(279.4930, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(124.0008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(227.9239, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(250.1208, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(302.5757, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(473.2384, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(255.8799, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(86.9942, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(109.3327, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(68.1604, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.6263, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.7657, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(107.5462, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(168.3831, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.20102651839178784, 'recall': 0.3092105263157895, 'f1': 0.24364955935717988, 'number': 760}, 'P': {'precision': 0.23658536585365852, 'recall': 0.23832923832923833, 'f1': 0.23745410036719705, 'number': 814}, 'overall_precision': 0.21568627450980393, 'overall_recall': 0.2725540025412961, 'overall_f1': 0.24080830760595004, 'overall_accuracy': 0.646122763533357}
			------------EPOCH 15---------------
Loss:  tensor(215.4778, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(200.0143, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(296.1655, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(144.9540, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(280.1476, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(97.7518, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(247.1999, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(216.8947, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(156.5760, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(254.7580, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(187.6227, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(590.7998, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(79.1170, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(205.1693, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(138.9422, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(211.8384, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(208.6048, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(84.1449, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(85.5965, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(88.7100, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(353.8977, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(337.0024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(233.5403, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(110.2433, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(311.6459, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(210.2059, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(231.2290, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(512.5656, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(425.5887, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(124.5645, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(198.9633, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(205.8901, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(89.6796, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(144.9750, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(464.5000, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(713.2476, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.16344916344916344, 'recall': 0.16710526315789473, 'f1': 0.16525699414443723, 'number': 760}, 'P': {'precision': 0.07692307692307693, 'recall': 0.14987714987714987, 'f1': 0.10166666666666667, 'number': 814}, 'overall_precision': 0.10537452391028354, 'overall_recall': 0.15819567979669633, 'overall_f1': 0.12649225298450598, 'overall_accuracy': 0.6587240424069484}
			------------EPOCH 16---------------
Loss:  tensor(385.5103, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(227.6967, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(304.3190, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(110.1780, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(171.7746, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.9703, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(155.3424, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.2300, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(91.5512, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(127.6421, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(77.9939, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(214.4917, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(72.8071, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(182.5647, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(127.3524, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(153.0047, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(622.1486, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(193.2027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(297.7256, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(187.7330, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(602.3783, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(696.2072, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(445.3184, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(205.1318, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(269.3986, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(293.9067, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(241.0498, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(394.6902, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(292.2556, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(67.9865, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(117.6607, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(113.6857, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.4280, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.2040, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(118.7206, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(166.5546, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.1937984496124031, 'recall': 0.09868421052631579, 'f1': 0.1307759372275501, 'number': 760}, 'P': {'precision': 0.10106716886377903, 'recall': 0.1977886977886978, 'f1': 0.13377648525135025, 'number': 814}, 'overall_precision': 0.1191919191919192, 'overall_recall': 0.14993646759847523, 'overall_f1': 0.1328081035453011, 'overall_accuracy': 0.6554539584678511}
			------------EPOCH 17---------------
Loss:  tensor(251.8851, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(192.3135, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(301.9861, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(139.6617, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(319.6587, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(265.8862, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(537.3452, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(217.8739, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(205.3391, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(450.0155, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(559.2679, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(707.4036, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.6917, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(221.1360, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(161.6435, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(252.1524, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(284.4402, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(76.7750, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(94.5391, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(96.4814, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(317.9849, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(301.2972, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(254.9099, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(132.8495, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(212.9569, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(203.7835, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(240.6073, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(289.1741, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(366.3093, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(84.5655, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(109.2550, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(91.4358, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.5026, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(69.6218, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(157.8142, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(196.5055, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.20078740157480315, 'recall': 0.3355263157894737, 'f1': 0.2512315270935961, 'number': 760}, 'P': {'precision': 0.15392354124748492, 'recall': 0.18796068796068796, 'f1': 0.16924778761061948, 'number': 814}, 'overall_precision': 0.18021201413427562, 'overall_recall': 0.25921219822109276, 'overall_f1': 0.21261073475768633, 'overall_accuracy': 0.6646810106017371}
			------------EPOCH 18---------------
Loss:  tensor(163.2900, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(152.3660, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(235.1436, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(74.8770, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(145.0674, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(59.1859, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(147.7080, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(121.3098, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(106.0618, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(241.1041, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(119.6555, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(239.4770, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.7863, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(143.4164, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(102.0183, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(106.8269, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(148.1884, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(75.2399, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(88.7604, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.3178, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(245.5745, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(287.7360, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(162.5298, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(77.2339, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(125.5172, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(117.4179, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(142.0136, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(221.0508, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(221.1994, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.3536, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(70.8428, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.1977, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.1957, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.9862, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(92.4751, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(111.6390, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.2134453781512605, 'recall': 0.16710526315789473, 'f1': 0.18745387453874537, 'number': 760}, 'P': {'precision': 0.13465346534653466, 'recall': 0.25061425061425063, 'f1': 0.17518248175182483, 'number': 814}, 'overall_precision': 0.15687203791469195, 'overall_recall': 0.210292249047014, 'overall_f1': 0.17969598262757874, 'overall_accuracy': 0.6786778030034784}
			------------EPOCH 19---------------
Loss:  tensor(95.7533, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(68.5580, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(148.7447, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.1259, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(113.9816, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.2112, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(127.5540, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(82.7231, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.4864, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(90.4128, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(139.1245, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(237.3489, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.5848, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(131.7976, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(65.7806, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(107.4770, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(120.3314, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.2227, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.4610, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.6238, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(220.4713, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(220.1153, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(138.3629, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.3774, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(115.6028, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(83.5604, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(116.8317, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(200.7134, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(165.0171, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.2201, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(65.3129, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.0222, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.2052, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.9652, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(82.2680, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(108.5741, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.16174055829228243, 'recall': 0.25921052631578945, 'f1': 0.19919110212335692, 'number': 760}, 'P': {'precision': 0.12825278810408922, 'recall': 0.16953316953316952, 'f1': 0.14603174603174604, 'number': 814}, 'overall_precision': 0.14603312990409764, 'overall_recall': 0.2128335451080051, 'overall_f1': 0.1732161323681489, 'overall_accuracy': 0.6699089791922684}
			------------EPOCH 20---------------
Loss:  tensor(83.6219, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(67.9830, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(100.9267, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.8792, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(65.8244, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.6427, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(61.4453, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.4868, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.0234, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(76.5848, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.7101, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(143.0557, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.8185, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(87.4672, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.3544, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.3571, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(66.3293, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.3290, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.2646, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.0692, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(194.7010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(197.0570, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(154.4665, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.8861, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(151.9723, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(85.2135, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(109.2641, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(167.2721, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(124.6751, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.6185, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.0093, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.5237, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.8569, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.0333, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(91.8413, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(71.5534, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.1959544879898862, 'recall': 0.20394736842105263, 'f1': 0.19987105093488072, 'number': 760}, 'P': {'precision': 0.14265025343953658, 'recall': 0.24201474201474202, 'f1': 0.17949886104783602, 'number': 814}, 'overall_precision': 0.16206261510128914, 'overall_recall': 0.22363405336721728, 'overall_f1': 0.18793379604911906, 'overall_accuracy': 0.6802816021328446}
			------------EPOCH 21---------------
Loss:  tensor(53.0362, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.8879, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.9166, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.0092, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.0107, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.9950, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.1305, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.9563, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.4099, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.5400, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.3236, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(103.7661, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.4553, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(70.4912, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.3540, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.5095, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.8355, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.9175, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.2978, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.1397, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(193.5007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(147.3119, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(121.3812, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.6460, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(105.4154, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(75.3304, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(120.4013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(145.7860, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(112.6446, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.2398, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.7979, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.8356, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.1612, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.5646, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.6754, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.1895, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.18008948545861297, 'recall': 0.2118421052631579, 'f1': 0.19467956469165656, 'number': 760}, 'P': {'precision': 0.13529856386999245, 'recall': 0.2199017199017199, 'f1': 0.16752456715021058, 'number': 814}, 'overall_precision': 0.15336039693279205, 'overall_recall': 0.21601016518424396, 'overall_f1': 0.17937219730941703, 'overall_accuracy': 0.687404969694445}
			------------EPOCH 22---------------
Loss:  tensor(40.9896, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.4764, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.8439, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.1483, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.1980, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.4543, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.2723, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.9190, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.7515, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.5719, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.9485, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(84.7873, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.3917, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.7128, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.6247, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.8919, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.5256, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.5905, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.2053, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.0997, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(119.0839, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(119.4787, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(91.6868, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.0351, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.5227, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.9977, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(90.2710, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(107.3830, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(103.4252, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.8431, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.8112, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.5664, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.1337, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.6312, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(69.4583, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.4443, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.185589519650655, 'recall': 0.2236842105263158, 'f1': 0.2028639618138425, 'number': 760}, 'P': {'precision': 0.17140600315955767, 'recall': 0.2665847665847666, 'f1': 0.20865384615384616, 'number': 814}, 'overall_precision': 0.1773602199816682, 'overall_recall': 0.24587039390088947, 'overall_f1': 0.20607028753993611, 'overall_accuracy': 0.6845931140780238}
			------------EPOCH 23---------------
Loss:  tensor(30.5093, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.5245, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.1397, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.0039, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.7285, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.6980, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.5775, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.0921, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.9590, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.0884, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.9708, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(69.2921, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.1449, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.1034, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.7320, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.2351, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.7403, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.8716, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.6639, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.2873, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(95.2714, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(97.3744, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(61.0522, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.6253, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.7673, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.2718, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(70.2381, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(98.1503, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(71.7246, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.9112, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.0334, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.4689, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.0646, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.8269, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(59.0184, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.7561, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.18040621266427717, 'recall': 0.1986842105263158, 'f1': 0.18910457107075768, 'number': 760}, 'P': {'precision': 0.12472647702407003, 'recall': 0.21007371007371006, 'f1': 0.15652173913043477, 'number': 814}, 'overall_precision': 0.14583333333333334, 'overall_recall': 0.204574332909784, 'overall_f1': 0.17028027498677947, 'overall_accuracy': 0.6840099143946179}
			------------EPOCH 24---------------
Loss:  tensor(27.9213, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.7578, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.6165, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.2735, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.7679, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.1992, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.9081, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.4306, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.4618, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.4620, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.3845, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(59.4129, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.6231, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(62.0836, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.7145, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.8872, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.9173, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.5432, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.9104, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.0595, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(89.8223, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(84.9105, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.4992, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.4364, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.0038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.9216, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.1941, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(79.7007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(62.9574, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.5788, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.5007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.4876, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.9432, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.9294, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.5014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.5410, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.19557195571955718, 'recall': 0.20921052631578949, 'f1': 0.20216147488874764, 'number': 760}, 'P': {'precision': 0.16767371601208458, 'recall': 0.2727272727272727, 'f1': 0.20767072029934516, 'number': 814}, 'overall_precision': 0.17828731867103417, 'overall_recall': 0.2420584498094028, 'overall_f1': 0.20533548908649962, 'overall_accuracy': 0.6828226864676845}
			------------EPOCH 25---------------
Loss:  tensor(14.1002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.1195, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.8060, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.4250, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.6537, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.7625, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.2085, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.2382, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.4576, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.2942, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.4439, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.9986, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.1121, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(94.5973, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.6504, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.5812, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.3193, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.5907, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.9181, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.7883, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(80.0131, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(76.2052, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.9592, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.8683, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.6255, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.1440, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.4277, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(59.1390, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.1924, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.6451, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.5224, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.3424, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.8916, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.3794, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.5877, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.7627, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.1814207650273224, 'recall': 0.21842105263157896, 'f1': 0.1982089552238806, 'number': 760}, 'P': {'precision': 0.15076923076923077, 'recall': 0.24078624078624078, 'f1': 0.18543046357615894, 'number': 814}, 'overall_precision': 0.16343115124153498, 'overall_recall': 0.22998729351969505, 'overall_f1': 0.1910794404856163, 'overall_accuracy': 0.6843223427964424}
			------------EPOCH 26---------------
Loss:  tensor(8.8097, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.0921, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.9937, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.0691, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.1168, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.0693, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.5821, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.2838, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.9119, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.2602, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.1465, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.9714, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.7139, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.5473, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.2373, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.0122, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.6879, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.7923, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.7371, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.3112, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(87.3841, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(66.0279, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.4835, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.8559, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.8984, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.2019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.9750, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.9229, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.0045, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.6397, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.1527, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.7574, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.3561, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.7804, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.0558, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.4705, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.20522388059701493, 'recall': 0.21710526315789475, 'f1': 0.21099744245524296, 'number': 760}, 'P': {'precision': 0.16654275092936802, 'recall': 0.2751842751842752, 'f1': 0.20750347383047704, 'number': 814}, 'overall_precision': 0.1810144253140996, 'overall_recall': 0.247141041931385, 'overall_f1': 0.20897125973677141, 'overall_accuracy': 0.6833017433504822}
			------------EPOCH 27---------------
Loss:  tensor(8.7139, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.3986, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.1512, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.6260, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.0592, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.2844, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.1904, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.1823, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.6954, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.2874, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.8170, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(61.8686, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.1120, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.6332, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.9952, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.7826, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.4897, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.7365, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.9092, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.8739, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(123.2641, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(71.1728, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.8242, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.3223, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.7191, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.3109, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(66.5242, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(87.1018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(66.7741, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.8304, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.7094, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.3988, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.7655, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.5457, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.8436, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.0017, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.18250539956803455, 'recall': 0.22236842105263158, 'f1': 0.20047449584816132, 'number': 760}, 'P': {'precision': 0.15023474178403756, 'recall': 0.23587223587223588, 'f1': 0.1835564053537285, 'number': 814}, 'overall_precision': 0.16379310344827586, 'overall_recall': 0.22935196950444728, 'overall_f1': 0.1911064055055585, 'overall_accuracy': 0.684405657036929}
			------------EPOCH 28---------------
Loss:  tensor(7.6183, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.5781, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.1881, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.1438, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.8860, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.2526, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.1933, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.0346, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.3115, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.8778, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.1445, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(117.1750, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.1999, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.7561, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.1007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.2690, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.0759, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.2926, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.4433, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.9721, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(145.8669, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(59.2492, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(70.2336, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.7066, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(61.0365, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.2422, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(64.1408, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(106.7662, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(146.9828, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.4050, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.6800, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.5675, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.8508, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.4324, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.8311, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.3335, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.24151967435549526, 'recall': 0.23421052631578948, 'f1': 0.23780895123580495, 'number': 760}, 'P': {'precision': 0.1549935149156939, 'recall': 0.2936117936117936, 'f1': 0.2028862478777589, 'number': 814}, 'overall_precision': 0.18297498903027642, 'overall_recall': 0.26493011435832275, 'overall_f1': 0.2164547106151051, 'overall_accuracy': 0.6833225719106039}
			------------EPOCH 29---------------
Loss:  tensor(9.5832, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.8392, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.1631, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.1642, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.0670, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.7410, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.0086, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.1336, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.3155, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.1199, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.5481, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(122.5469, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.0678, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(128.8719, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.6999, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(91.7324, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.5663, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.4075, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.0157, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.7450, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(199.5072, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(92.5259, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(74.5923, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.3969, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(64.8268, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(73.4682, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(123.9076, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(157.8072, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(255.0433, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.4761, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.8819, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.7654, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.8140, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.9759, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.0785, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.8460, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.178714859437751, 'recall': 0.23421052631578948, 'f1': 0.20273348519362186, 'number': 760}, 'P': {'precision': 0.11360123647604328, 'recall': 0.18058968058968058, 'f1': 0.1394686907020873, 'number': 814}, 'overall_precision': 0.14192139737991266, 'overall_recall': 0.20648030495552733, 'overall_f1': 0.16821946169772256, 'overall_accuracy': 0.6845306283976589}
			------------EPOCH 30---------------
Loss:  tensor(22.1100, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.9107, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.9178, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.0319, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.2970, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.6536, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.8740, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.3288, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.2598, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.2132, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.6699, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(77.8946, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.4589, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(112.3400, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.0598, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(61.3948, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.7231, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.7487, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.5891, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.9572, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(141.0095, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(136.1951, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(66.8179, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.9664, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.9267, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(61.1200, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(100.7675, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(69.3246, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(124.9537, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.1386, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.1584, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.7636, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.8941, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.5232, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.9130, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.5248, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.2297142857142857, 'recall': 0.2644736842105263, 'f1': 0.24587155963302754, 'number': 760}, 'P': {'precision': 0.15823235923022094, 'recall': 0.2727272727272727, 'f1': 0.20027063599458728, 'number': 814}, 'overall_precision': 0.18568920105355574, 'overall_recall': 0.2687420584498094, 'overall_f1': 0.21962616822429906, 'overall_accuracy': 0.6887171689821082}


		-------------RUN 3-----------
			------------EPOCH 1---------------
Loss:  tensor(1802.0389, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1956.5476, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2240.8447, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2775.2258, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1067.7892, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2240.2590, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1336.4423, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1007.8318, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1970.5405, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2128.9045, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1596.8901, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1780.7778, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1915.4795, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1434.3499, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2827.0942, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1601.3926, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1935.9307, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1050.6234, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1468.9836, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1878.6155, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1578.6978, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1282.6058, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1652.4990, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2008.5072, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1088.1853, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1706.6650, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1688.6484, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1268.9666, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2995.2964, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2309.1731, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1506.9386, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1048.2629, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1995.7987, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1204.9683, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1203.0935, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1619.9351, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2277.4636, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.0787037037037037, 'recall': 0.025835866261398176, 'f1': 0.038901601830663615, 'number': 658}, 'P': {'precision': 0.2, 'recall': 0.26816239316239315, 'f1': 0.22911912368781379, 'number': 936}, 'overall_precision': 0.18218898708361658, 'overall_recall': 0.16813048933500627, 'overall_f1': 0.17487765089722676, 'overall_accuracy': 0.6109728814268092}
			------------EPOCH 2---------------
Loss:  tensor(1107.2510, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1203.1357, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1304.0886, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1742.3623, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(696.6638, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1523.2686, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(911.8448, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(645.7354, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1449.0479, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1416.0972, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1141.6506, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1414.3313, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1485.9641, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1068.0143, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2338.0288, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1459.7473, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1827.1306, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(909.7498, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1312.3265, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1776.6204, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1437.3284, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1163.1755, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1403.2083, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1705.4686, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(895.2067, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1468.4988, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1486.0781, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1071.2928, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2651.5381, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2001.1003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1162.1030, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(865.7406, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1675.0881, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1088.5078, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1020.7983, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1376.7864, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2033.4723, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.19402985074626866, 'recall': 0.0790273556231003, 'f1': 0.11231101511879048, 'number': 658}, 'P': {'precision': 0.188470066518847, 'recall': 0.2724358974358974, 'f1': 0.22280471821756226, 'number': 936}, 'overall_precision': 0.189389265885256, 'overall_recall': 0.19259723964868256, 'overall_f1': 0.19097978227060652, 'overall_accuracy': 0.6223472097350868}
			------------EPOCH 3---------------
Loss:  tensor(993.2222, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1092.7556, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1190.5073, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1592.6782, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(610.0543, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1257.1364, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(786.3722, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(577.0155, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1280.8425, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1293.4946, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1042.1799, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1052.0804, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1317.7480, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(856.7142, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1955.9229, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1252.7354, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1375.2607, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(720.4209, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(942.8016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1492.8611, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1244.1248, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(958.6040, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1185.6090, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1542.8304, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(775.9542, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1252.6278, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1289.1830, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(981.3611, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2334.6831, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1702.3250, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(914.6629, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(692.8662, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1461.7908, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1010.7892, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(843.5069, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1149.2229, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1719.0836, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.09396751740139211, 'recall': 0.12310030395136778, 'f1': 0.10657894736842106, 'number': 658}, 'P': {'precision': 0.12890995260663507, 'recall': 0.1452991452991453, 'f1': 0.1366147664490206, 'number': 936}, 'overall_precision': 0.11319770474700053, 'overall_recall': 0.13613550815558345, 'overall_f1': 0.12361150669324979, 'overall_accuracy': 0.639074163129613}
			------------EPOCH 4---------------
Loss:  tensor(838.9116, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(905.1539, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(904.8095, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1365.7175, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(508.2014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(995.0834, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(663.8591, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(470.0701, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1029.4595, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1072.3505, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(842.5415, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(839.3077, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1106.9351, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(618.4175, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1641.4589, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1084.4810, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(950.2645, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(521.8044, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(565.9357, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1254.0121, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1072.3005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(758.3326, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(941.4701, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1390.1307, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(646.7842, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(943.6390, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1089.6677, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(831.8541, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1915.3015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1294.1920, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(704.6271, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(523.0925, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1214.7615, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(900.3589, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(653.2641, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(924.2994, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1520.4414, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.10327455919395466, 'recall': 0.18693009118541035, 'f1': 0.13304488912925908, 'number': 658}, 'P': {'precision': 0.09090909090909091, 'recall': 0.06944444444444445, 'f1': 0.07874015748031497, 'number': 936}, 'overall_precision': 0.09863588667366212, 'overall_recall': 0.11794228356336262, 'overall_f1': 0.10742857142857141, 'overall_accuracy': 0.6186254626048048}
			------------EPOCH 5---------------
Loss:  tensor(695.8755, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(779.0795, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(721.6910, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1195.1182, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(425.9731, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(793.0526, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(541.5187, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(383.4781, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(879.3577, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(900.8359, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(679.4331, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(731.7460, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(958.9069, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(433.6485, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1303.4885, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(907.1527, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(636.8186, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(398.5497, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(339.0214, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1014.5796, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(842.9969, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(639.8597, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(785.1014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1260.9579, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(607.7521, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(783.2222, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(926.1243, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(728.0157, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1769.5225, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1069.9634, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(584.8232, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(415.7502, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1006.0875, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(802.1390, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(542.7039, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(772.8928, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1160.8362, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.1173974540311174, 'recall': 0.25227963525835867, 'f1': 0.16023166023166024, 'number': 658}, 'P': {'precision': 0.2563667232597623, 'recall': 0.16132478632478633, 'f1': 0.19803278688524592, 'number': 936}, 'overall_precision': 0.1582626060908637, 'overall_recall': 0.198870765370138, 'overall_f1': 0.17625799277175422, 'overall_accuracy': 0.6216781315993058}
			------------EPOCH 6---------------
Loss:  tensor(582.7094, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(676.1992, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(554.6599, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1000.2347, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(352.3001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(663.1559, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(444.6611, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(335.3641, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(747.5280, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(727.0909, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(585.5994, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(579.3250, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(925.6243, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(395.2798, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1215.8210, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(828.4056, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(543.4046, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(330.3206, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(262.6228, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(857.9492, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(609.4653, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(489.0965, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(647.8045, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1029.5132, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(527.5138, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(647.1122, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(718.1653, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(592.2167, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1542.8088, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(806.7273, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(443.8504, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(308.8813, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(824.1964, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(735.4058, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(445.1631, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(676.6661, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1037.5750, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.14831905075807514, 'recall': 0.34194528875379937, 'f1': 0.20689655172413793, 'number': 658}, 'P': {'precision': 0.3249551166965889, 'recall': 0.19337606837606838, 'f1': 0.24246483590087076, 'number': 936}, 'overall_precision': 0.19575699132111862, 'overall_recall': 0.2547051442910916, 'overall_f1': 0.22137404580152673, 'overall_accuracy': 0.584439751604742}
			------------EPOCH 7---------------
Loss:  tensor(544.5534, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(680.2336, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(531.4006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(977.2832, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(273.0023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(722.8081, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(409.9164, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(344.9755, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(865.8049, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(687.0585, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(524.2292, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(569.7606, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(724.0288, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(252.7700, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1167.2372, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(638.7515, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(337.1291, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(227.4739, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(171.8028, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(621.3666, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(537.4874, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(431.3248, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(570.8850, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(923.9340, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(541.8472, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(495.7103, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(677.5143, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(622.9348, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1499.6157, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(847.4875, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(502.2284, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(353.8283, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(726.2346, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(637.7900, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(516.5452, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(757.9365, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(860.0906, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.14733542319749215, 'recall': 0.21428571428571427, 'f1': 0.1746130030959752, 'number': 658}, 'P': {'precision': 0.22174288179465057, 'recall': 0.2745726495726496, 'f1': 0.24534606205250598, 'number': 936}, 'overall_precision': 0.1880907372400756, 'overall_recall': 0.24968632371392724, 'overall_f1': 0.21455525606469003, 'overall_accuracy': 0.6692454053149894}
			------------EPOCH 8---------------
Loss:  tensor(408.0775, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(511.9683, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(437.9887, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(938.4845, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(219.8899, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(363.0058, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(263.1452, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(265.2278, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(525.0876, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(518.7935, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(398.7351, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(492.0612, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(752.7722, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(271.2875, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1234.6277, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(819.4674, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(555.6780, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(250.1997, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(182.7791, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(796.2079, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(678.5093, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(696.0525, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(771.7256, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(998.9707, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(403.9557, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(575.9505, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(590.9819, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(468.7668, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1040.5503, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(555.3348, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(243.0205, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(208.1729, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(493.4614, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(529.0106, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(303.2503, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(494.0646, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(754.0966, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.2773722627737226, 'recall': 0.11550151975683891, 'f1': 0.1630901287553648, 'number': 658}, 'P': {'precision': 0.23133826496301277, 'recall': 0.36752136752136755, 'f1': 0.283945522080066, 'number': 936}, 'overall_precision': 0.23850085178875638, 'overall_recall': 0.26348808030112925, 'overall_f1': 0.2503725782414307, 'overall_accuracy': 0.6581847073828591}
			------------EPOCH 9---------------
Loss:  tensor(478.1216, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(634.1136, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(660.7632, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1214.8850, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(272.9442, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(540.3265, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(323.9300, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(376.5877, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(681.3171, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(750.2253, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(560.2366, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(295.0957, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(740.2139, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(231.9387, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(835.2323, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(517.5900, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(272.5545, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(212.0796, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(141.9258, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(498.5017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(485.2399, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(303.5848, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(412.4608, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(695.3198, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(406.1754, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(389.2136, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(410.3687, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(428.4141, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1147.4625, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(561.4612, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(273.4557, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(255.0601, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(584.0099, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(522.1921, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(331.8223, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(474.6319, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(596.9245, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.16177777777777777, 'recall': 0.2765957446808511, 'f1': 0.20415030846887272, 'number': 658}, 'P': {'precision': 0.26871794871794874, 'recall': 0.2799145299145299, 'f1': 0.2742019884877028, 'number': 936}, 'overall_precision': 0.21142857142857144, 'overall_recall': 0.27854454203262236, 'overall_f1': 0.24038982133188957, 'overall_accuracy': 0.669851757375541}
			------------EPOCH 10---------------
Loss:  tensor(278.5452, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(312.7729, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(265.5878, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(614.0201, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(167.0289, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(218.9780, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(175.6675, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(220.4486, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(386.2375, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(394.1003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(286.4293, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(264.9590, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(582.1769, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(147.2772, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(806.9598, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(396.3841, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(245.5726, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(169.9397, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(93.2357, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(458.3678, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(421.3897, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(335.0109, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(376.8818, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(857.0750, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(303.6235, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(440.4862, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(344.3455, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(423.3638, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(729.7510, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(428.0035, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(146.2376, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(119.6812, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(363.5185, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(372.7117, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(184.8534, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(304.6315, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(433.2960, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.12202609363008442, 'recall': 0.24164133738601823, 'f1': 0.16216216216216217, 'number': 658}, 'P': {'precision': 0.10486486486486486, 'recall': 0.10363247863247864, 'f1': 0.10424502955400322, 'number': 936}, 'overall_precision': 0.11490125673249552, 'overall_recall': 0.1606022584692597, 'overall_f1': 0.13396127681841968, 'overall_accuracy': 0.6491730612415582}
			------------EPOCH 11---------------
Loss:  tensor(243.2888, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(288.7275, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(291.3375, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(598.0416, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(157.8584, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(242.7563, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(154.6527, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(198.0428, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(317.4003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(316.2073, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(263.4480, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(273.4392, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(446.1611, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(93.6012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(473.3573, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(280.0583, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(130.2806, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(95.7509, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.9716, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(292.0787, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(278.5930, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(169.5330, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(224.3528, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(465.2021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(240.1346, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(254.0009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(206.5268, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(322.4055, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(672.6722, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(301.3057, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(97.2595, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(92.6474, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(222.5128, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(257.2080, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(150.1564, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(282.5140, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(342.5093, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.19154228855721392, 'recall': 0.23404255319148937, 'f1': 0.21067031463748292, 'number': 658}, 'P': {'precision': 0.26666666666666666, 'recall': 0.32905982905982906, 'f1': 0.29459588713534196, 'number': 936}, 'overall_precision': 0.23583460949464014, 'overall_recall': 0.28983688833124216, 'overall_f1': 0.26006191950464397, 'overall_accuracy': 0.6807870031572124}
			------------EPOCH 12---------------
Loss:  tensor(181.7870, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(211.6645, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(174.7251, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(500.1899, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(97.0431, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(107.3750, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(72.3851, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(132.4759, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(198.3929, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(226.6586, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(181.0843, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(134.6656, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(325.1742, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.4764, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(332.4852, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(209.0866, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(88.7408, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(74.8443, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.1417, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(223.2915, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(208.4675, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(118.0421, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(155.0206, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(352.4353, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(174.0493, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(197.7568, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(153.3741, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(255.2784, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(496.8293, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(225.0947, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(65.9908, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(69.8096, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(170.5583, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(177.5798, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(94.1275, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(189.9847, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(212.8329, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.15648148148148147, 'recall': 0.256838905775076, 'f1': 0.19447640966628307, 'number': 658}, 'P': {'precision': 0.18521983161833488, 'recall': 0.21153846153846154, 'f1': 0.19750623441396506, 'number': 936}, 'overall_precision': 0.17077710563052584, 'overall_recall': 0.2302383939774153, 'overall_f1': 0.19609938551963665, 'overall_accuracy': 0.6719217178581136}
			------------EPOCH 13---------------
Loss:  tensor(119.4523, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(134.3382, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(116.7546, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(307.9438, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(67.6332, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(76.0070, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.8237, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(106.0514, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(135.0819, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(138.0210, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(109.1904, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(96.5350, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(249.2032, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.4739, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(246.5216, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(150.7430, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(64.9379, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.0013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.2130, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(165.6164, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(154.4534, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(86.0304, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(103.3179, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(305.0804, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(144.4359, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(195.6031, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(132.1272, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(209.8507, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(388.4326, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(192.2805, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.3595, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.3770, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(93.7381, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(118.6396, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.1689, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(134.5344, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(153.7931, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.16302367941712204, 'recall': 0.27203647416413373, 'f1': 0.20387243735763097, 'number': 658}, 'P': {'precision': 0.19869402985074627, 'recall': 0.22756410256410256, 'f1': 0.21215139442231076, 'number': 936}, 'overall_precision': 0.18064516129032257, 'overall_recall': 0.24592220828105396, 'overall_f1': 0.20828905419766205, 'overall_accuracy': 0.6715453614067368}
			------------EPOCH 14---------------
Loss:  tensor(83.6962, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(91.1757, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(73.7675, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(229.0335, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.3594, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(59.6673, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.7945, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(85.5717, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(124.5901, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(131.8610, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(96.2588, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(85.1295, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(198.9059, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.9348, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(203.1263, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(122.9924, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.6891, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.5350, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.2111, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(116.8571, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(105.9663, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(82.4962, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(69.2081, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(224.2737, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(104.1021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(157.6690, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(98.3275, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(128.4055, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(300.4052, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(118.7559, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.1681, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.9689, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(59.0597, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(75.5507, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.2941, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(96.3981, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(148.1361, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.17509247842170161, 'recall': 0.21580547112462006, 'f1': 0.1933287950987066, 'number': 658}, 'P': {'precision': 0.22083007047768208, 'recall': 0.30128205128205127, 'f1': 0.25485765928603704, 'number': 936}, 'overall_precision': 0.20306513409961685, 'overall_recall': 0.2659974905897114, 'overall_f1': 0.23030961434003258, 'overall_accuracy': 0.6752671085370189}
			------------EPOCH 15---------------
Loss:  tensor(72.4684, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(74.5945, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.9504, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(195.5691, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.5223, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.0447, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.3837, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(61.7554, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(94.6064, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(118.7723, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(80.5454, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(59.6165, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(153.3130, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.8115, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(139.7722, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(95.7802, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.3779, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.0538, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.9325, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(75.5945, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(72.7002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.9965, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.6795, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(170.4188, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(83.4018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(101.7284, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(66.5202, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(85.4023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(214.4212, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(88.3553, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.6969, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.0142, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.4786, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.0531, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.4360, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(76.0446, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(149.5178, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.15422885572139303, 'recall': 0.2826747720364742, 'f1': 0.19957081545064378, 'number': 658}, 'P': {'precision': 0.1757631822386679, 'recall': 0.202991452991453, 'f1': 0.18839861179970255, 'number': 936}, 'overall_precision': 0.1644075207695671, 'overall_recall': 0.23588456712672523, 'overall_f1': 0.193764493687194, 'overall_accuracy': 0.6674472578250779}
			------------EPOCH 16---------------
Loss:  tensor(50.5642, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.1458, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.5692, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(135.0587, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.8876, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.8819, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.3488, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.5744, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(78.7772, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(101.3563, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(65.1279, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.7241, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(134.4542, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.7285, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(106.0431, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(73.1429, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.9896, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.9993, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.4101, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.5440, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.0023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.6226, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.0558, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(138.8479, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(61.5810, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(76.7429, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.9999, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.7681, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(186.5687, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.3441, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.1598, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.5810, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.1188, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.3385, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.5485, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.9536, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(65.5427, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.17777777777777778, 'recall': 0.2553191489361702, 'f1': 0.2096069868995633, 'number': 658}, 'P': {'precision': 0.18911685994647637, 'recall': 0.2264957264957265, 'f1': 0.20612542537676226, 'number': 936}, 'overall_precision': 0.18393030009680542, 'overall_recall': 0.2383939774153074, 'overall_f1': 0.20765027322404372, 'overall_accuracy': 0.6748489347021557}
			------------EPOCH 17---------------
Loss:  tensor(36.5229, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.3944, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.0696, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(95.1556, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.6571, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.7544, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.4517, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.6124, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.7940, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.3259, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.3619, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.6299, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(93.7232, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.6144, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(95.4980, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(62.2635, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.0803, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.7509, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.3109, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.4531, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.8070, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.9497, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.7881, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(125.7590, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.4925, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(66.2311, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.2516, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.7411, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(137.6135, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.7621, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.2585, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.4386, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.5096, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.5004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.9156, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.2637, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.1194, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.18655097613882862, 'recall': 0.2613981762917933, 'f1': 0.21772151898734174, 'number': 658}, 'P': {'precision': 0.20435120435120435, 'recall': 0.280982905982906, 'f1': 0.23661718398560502, 'number': 936}, 'overall_precision': 0.19692168401991852, 'overall_recall': 0.2728983688833124, 'overall_f1': 0.22876676308177757, 'overall_accuracy': 0.6750371129278441}
			------------EPOCH 18---------------
Loss:  tensor(25.5322, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.5961, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.5689, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(85.0005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.2647, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.4769, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.7820, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.8705, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.9290, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.3759, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.9724, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.2583, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(84.9821, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.4756, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(81.9615, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.5026, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.1340, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.1419, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.0931, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.8474, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.3411, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.4331, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.1357, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(108.8684, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.5873, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(66.0819, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.2381, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.1369, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(118.7266, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(71.6834, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.5493, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.4996, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.9923, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.1183, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.9186, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.6361, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(93.7300, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.15069551777434312, 'recall': 0.29635258358662614, 'f1': 0.19979508196721307, 'number': 658}, 'P': {'precision': 0.183920704845815, 'recall': 0.17841880341880342, 'f1': 0.18112798264642083, 'number': 936}, 'overall_precision': 0.16439600363306087, 'overall_recall': 0.22710163111668757, 'overall_f1': 0.19072708113804007, 'overall_accuracy': 0.6580592552324}
			------------EPOCH 19---------------
Loss:  tensor(27.2884, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.8414, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.6370, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(77.4036, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.5414, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.1907, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.5679, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.0222, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.2848, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.1107, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.2647, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.5579, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(80.9919, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.9630, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(70.6008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(99.8250, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.3977, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.3132, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.9907, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.6042, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.1781, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.2194, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.2948, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(93.7055, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.4137, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.9659, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.7640, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.1798, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(100.2896, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.6806, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.4800, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.8517, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.3583, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.3638, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.9467, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.7355, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.7618, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.2076923076923077, 'recall': 0.24620060790273557, 'f1': 0.22531293463143257, 'number': 658}, 'P': {'precision': 0.18772563176895307, 'recall': 0.2777777777777778, 'f1': 0.22404136148211978, 'number': 936}, 'overall_precision': 0.19491916859122402, 'overall_recall': 0.26474278544542035, 'overall_f1': 0.22452779994679437, 'overall_accuracy': 0.6783197775315198}
			------------EPOCH 20---------------
Loss:  tensor(18.8661, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.1113, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.7056, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(66.9166, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.6198, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.3586, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.4756, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.9571, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.0247, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.5647, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.1621, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.0228, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(74.8168, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.9823, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.2885, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(65.1497, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.5876, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.8738, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.0813, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.0501, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.7321, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.5876, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.1777, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(78.7100, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.4242, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.1585, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.9735, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.8598, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(77.7110, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.5106, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.9803, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.5895, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.6882, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.2405, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.4964, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.3993, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.2134, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.1645133505598622, 'recall': 0.29027355623100304, 'f1': 0.21000549752611328, 'number': 658}, 'P': {'precision': 0.1829155060352832, 'recall': 0.21047008547008547, 'f1': 0.19572776949826132, 'number': 936}, 'overall_precision': 0.17336907953529937, 'overall_recall': 0.24341279799247176, 'overall_f1': 0.20250521920668055, 'overall_accuracy': 0.6687226880214105}
			------------EPOCH 21---------------
Loss:  tensor(8.9733, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.9360, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.0948, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.5504, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.7861, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.4852, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.2808, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.7113, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.4568, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.3543, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.9770, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.3386, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.5359, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.4808, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.5391, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.1115, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.9788, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.0637, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5784, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.8783, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.0762, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.3718, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.4941, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.1621, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.3640, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(67.4351, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.6474, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.8223, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(73.0856, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.2095, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.3702, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.2432, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.9567, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.2722, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.9802, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.5790, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.8899, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.17788944723618091, 'recall': 0.2689969604863222, 'f1': 0.21415607985480947, 'number': 658}, 'P': {'precision': 0.17047308319738988, 'recall': 0.2232905982905983, 'f1': 0.19333950046253467, 'number': 936}, 'overall_precision': 0.17379558757316524, 'overall_recall': 0.24215809284818068, 'overall_f1': 0.20235910878112714, 'overall_accuracy': 0.677943421080143}
			------------EPOCH 22---------------
Loss:  tensor(5.6870, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.4999, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.6943, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.7743, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.0684, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.6673, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.3440, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.9926, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.9546, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.5331, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.8486, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.0579, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.6163, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.4639, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.4365, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.4677, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.7200, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.3974, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.1405, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.6947, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.5261, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.2112, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.8593, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.3201, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.0685, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.0822, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.8338, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.8069, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.9070, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.5500, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.3415, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5042, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.9789, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.7369, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.6837, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.7929, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.0613, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.1724461105904405, 'recall': 0.2796352583586626, 'f1': 0.2133333333333333, 'number': 658}, 'P': {'precision': 0.20506108202443282, 'recall': 0.25106837606837606, 'f1': 0.22574447646493753, 'number': 936}, 'overall_precision': 0.18933574333483957, 'overall_recall': 0.26286072772898367, 'overall_f1': 0.22012083004990807, 'overall_accuracy': 0.6724026177682062}
			------------EPOCH 23---------------
Loss:  tensor(4.5819, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.3498, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.0601, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.7188, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.3391, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.0091, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5935, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.0318, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.1786, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.9989, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.7776, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.1653, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.2173, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.9472, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.8297, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.3755, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.6856, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.7350, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.9824, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.8380, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.9973, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.5886, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.1664, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.8032, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.9376, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.8450, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.4346, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.3823, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.5000, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.2564, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.4930, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.4778, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.6350, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.5798, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.8763, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.3367, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.7820, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.1697381183317168, 'recall': 0.26595744680851063, 'f1': 0.2072232089994079, 'number': 658}, 'P': {'precision': 0.1723027375201288, 'recall': 0.22863247863247863, 'f1': 0.19651056014692378, 'number': 936}, 'overall_precision': 0.17113946326440826, 'overall_recall': 0.2440401505646173, 'overall_f1': 0.20118955262477373, 'overall_accuracy': 0.6760407301315157}
			------------EPOCH 24---------------
Loss:  tensor(3.4258, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.0605, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.3883, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.5413, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.7313, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.6105, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.3604, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.6433, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.5860, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.7602, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.0289, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.8595, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.5462, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.4994, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.9187, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.3277, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.1193, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.8013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5854, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.7328, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.6760, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.5647, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.1681, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.4962, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.4561, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.3555, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.8283, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.5415, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.3506, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.9179, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.0074, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.8371, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.4957, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.8448, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.0660, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.0355, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.2461, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.17006200177147918, 'recall': 0.2917933130699088, 'f1': 0.2148852825965305, 'number': 658}, 'P': {'precision': 0.18699910952804988, 'recall': 0.22435897435897437, 'f1': 0.20398251578436136, 'number': 936}, 'overall_precision': 0.17850799289520428, 'overall_recall': 0.2521957340025094, 'overall_f1': 0.20904836193447737, 'overall_accuracy': 0.668534509795722}
			------------EPOCH 25---------------
Loss:  tensor(3.0806, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.4773, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.4009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.4418, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2856, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.0565, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.7640, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.9063, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.5773, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.3252, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.1497, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.4200, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.2856, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2708, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.0604, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.7866, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.7305, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.8646, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3268, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.1684, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.9387, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.8822, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.9155, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.4757, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.3913, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.7419, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.4988, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.1755, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.5767, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.3600, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.7848, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6822, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.0512, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.2162, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.8005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.9498, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.2946, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.1719056974459725, 'recall': 0.26595744680851063, 'f1': 0.20883054892601433, 'number': 658}, 'P': {'precision': 0.17459080280592362, 'recall': 0.23931623931623933, 'f1': 0.20189274447949526, 'number': 936}, 'overall_precision': 0.17340286831812254, 'overall_recall': 0.25031367628607276, 'overall_f1': 0.2048780487804878, 'overall_accuracy': 0.6747652999351831}
			------------EPOCH 26---------------
Loss:  tensor(2.5960, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.8228, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.2013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.9105, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.0156, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.1372, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5931, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.0055, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.3448, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.9092, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.8976, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.5892, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.2325, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.1118, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.5008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.0286, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.7983, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.0426, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3492, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.5285, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.3919, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.7073, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.2307, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.8501, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.1698, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(82.0445, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.9239, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.5776, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(93.5544, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.3352, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6850, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6553, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.2109, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.0173, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.0398, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.3150, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.4405, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.16417910447761194, 'recall': 0.28419452887537994, 'f1': 0.20812465219810797, 'number': 658}, 'P': {'precision': 0.19944082013047532, 'recall': 0.22863247863247863, 'f1': 0.21304131408661026, 'number': 936}, 'overall_precision': 0.18128390596745028, 'overall_recall': 0.25156838143036386, 'overall_f1': 0.21071991592222808, 'overall_accuracy': 0.6679908838104}
			------------EPOCH 27---------------
Loss:  tensor(2.4838, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.4197, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5575, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.5079, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.9159, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.9788, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5108, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6624, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.0944, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.2436, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.6345, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.3743, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.3525, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.8318, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.2447, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.6895, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.4311, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.7488, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2028, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.4019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.7413, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.2612, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.5202, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.2166, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.3591, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(69.5210, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.3828, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.4925, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(80.9927, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.4103, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.7320, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5341, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.9269, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.9590, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.4018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.6076, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(59.2595, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.17873303167420815, 'recall': 0.24012158054711247, 'f1': 0.20492866407263297, 'number': 658}, 'P': {'precision': 0.17037037037037037, 'recall': 0.24572649572649571, 'f1': 0.20122484689413822, 'number': 936}, 'overall_precision': 0.17367949865711726, 'overall_recall': 0.24341279799247176, 'overall_f1': 0.20271682340647856, 'overall_accuracy': 0.6761870909737178}
			------------EPOCH 28---------------
Loss:  tensor(2.3442, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.5979, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.0359, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.2153, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.7487, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.7974, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2964, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6459, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.7547, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.6563, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.2461, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.1970, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.8836, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6030, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.1840, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.9111, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.8363, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.6756, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1700, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.7831, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.1163, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.9659, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.4722, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(77.2697, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.4091, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(82.4335, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.6260, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.8664, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(72.0455, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.3544, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.8486, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4414, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.1220, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.1888, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.5022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.4140, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(105.4787, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.16607142857142856, 'recall': 0.2826747720364742, 'f1': 0.20922384701912264, 'number': 658}, 'P': {'precision': 0.21766848816029144, 'recall': 0.2553418803418803, 'f1': 0.23500491642084562, 'number': 936}, 'overall_precision': 0.19161406672678089, 'overall_recall': 0.266624843161857, 'overall_f1': 0.2229800629590766, 'overall_accuracy': 0.6713990005645347}
			------------EPOCH 29---------------
Loss:  tensor(2.9791, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.8384, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.2724, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(61.1159, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.9159, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.4828, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3954, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5754, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.2385, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.9670, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.9661, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.9431, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.8584, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6470, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.8361, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.7053, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.0350, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.9797, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0277, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.0209, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.9791, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.6797, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.9909, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.5158, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.6682, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.1230, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.7698, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.7729, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(65.2154, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.6983, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.7120, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5501, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.7406, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.2233, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5435, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.2946, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.3304, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.1769406392694064, 'recall': 0.23556231003039513, 'f1': 0.20208604954367665, 'number': 658}, 'P': {'precision': 0.16791044776119404, 'recall': 0.2403846153846154, 'f1': 0.19771528998242532, 'number': 936}, 'overall_precision': 0.17148014440433212, 'overall_recall': 0.2383939774153074, 'overall_f1': 0.19947506561679787, 'overall_accuracy': 0.6788215861333556}
			------------EPOCH 30---------------
Loss:  tensor(3.0303, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.5460, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.6969, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.4613, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.8161, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.7944, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4709, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.8057, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.0056, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.3412, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.3132, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.3696, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.8456, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5207, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.5342, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.6831, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.8814, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.7715, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9706, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.5433, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.3388, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.9553, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.5825, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.6222, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.7686, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(64.3428, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.1882, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.6258, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.7931, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.6748, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4924, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2518, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.3216, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5219, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.1189, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.7334, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.7680, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.1771117166212534, 'recall': 0.29635258358662614, 'f1': 0.22171688459351907, 'number': 658}, 'P': {'precision': 0.20344827586206896, 'recall': 0.25213675213675213, 'f1': 0.22519083969465647, 'number': 936}, 'overall_precision': 0.19062361786819992, 'overall_recall': 0.2703889585947302, 'overall_f1': 0.22360570687418935, 'overall_accuracy': 0.6742007652581178}


		-------------RUN 4-----------
			------------EPOCH 1---------------
Loss:  tensor(1856.0142, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2915.2490, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3703.9995, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3486.8574, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2460.0298, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2197.1621, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1780.1426, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2036.4973, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1476.6261, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1122.4221, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2585.0630, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2700.2556, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1987.4309, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1295.6543, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(894.2960, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1063.4530, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1913.2797, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1684.1230, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1631.9434, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1420.3215, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1953.8938, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1518.1038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1490.8087, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2484.6416, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1387.1628, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2228.9626, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1665.3274, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1779.6342, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1561.3022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1378.3164, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1380.3867, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1418.6780, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2041.0754, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(969.9799, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1683.0785, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3522.4209, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2033.7468, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1663.9755, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.04551724137931035, 'recall': 0.04551724137931035, 'f1': 0.04551724137931035, 'number': 725}, 'P': {'precision': 0.08419838523644751, 'recall': 0.09193954659949623, 'f1': 0.08789885611077665, 'number': 794}, 'overall_precision': 0.06658291457286432, 'overall_recall': 0.06978275181040158, 'overall_f1': 0.06814529090324653, 'overall_accuracy': 0.6108508454914559}
			------------EPOCH 2---------------
Loss:  tensor(1124.8085, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1976.3698, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2508.3936, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2357.8870, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2119.8682, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1516.4280, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1232.5902, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1667.8047, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1037.1965, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(739.1124, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1924.6785, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2036.4613, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1553.2688, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(957.4841, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(690.4114, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(843.1885, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1468.8015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1467.9656, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1275.9762, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1146.4883, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1650.5992, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1350.8080, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1372.9727, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1981.6908, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1089.5775, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1613.1820, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1401.1628, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1570.8833, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1349.8938, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1198.2103, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1203.4174, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1231.0879, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1762.8843, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(844.1472, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1433.5608, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3359.8999, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1796.5732, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1456.4196, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.029082774049217, 'recall': 0.01793103448275862, 'f1': 0.02218430034129693, 'number': 725}, 'P': {'precision': 0.10052910052910052, 'recall': 0.14357682619647355, 'f1': 0.11825726141078838, 'number': 794}, 'overall_precision': 0.08032890575585072, 'overall_recall': 0.0836076366030283, 'overall_f1': 0.08193548387096775, 'overall_accuracy': 0.6117878017222148}
			------------EPOCH 3---------------
Loss:  tensor(997.4026, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1658.5603, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2074.5698, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1973.3339, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1894.9644, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1296.1216, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1004.0427, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1360.3772, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(903.6541, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(626.6607, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1594.7559, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1753.0542, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1342.5923, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(831.4334, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(604.7432, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(730.2035, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1270.6810, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1313.3911, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1110.4915, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(996.7856, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1405.1396, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1155.2144, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1216.6573, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1793.3740, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(935.9541, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1329.1217, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1156.3345, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1413.8086, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1143.4314, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1060.9888, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1046.2919, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1031.7706, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1559.5415, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(757.3323, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1225.8197, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2954.4963, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1533.0641, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1350.4474, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.07644110275689223, 'recall': 0.08413793103448276, 'f1': 0.08010505581089954, 'number': 725}, 'P': {'precision': 0.09054325955734406, 'recall': 0.11335012594458438, 'f1': 0.10067114093959732, 'number': 794}, 'overall_precision': 0.08426339285714286, 'overall_recall': 0.09940750493745885, 'overall_f1': 0.09121111446692841, 'overall_accuracy': 0.6153125418284031}
			------------EPOCH 4---------------
Loss:  tensor(871.4739, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1422.5403, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1813.7532, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1793.0359, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1624.1304, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1085.4745, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(820.3092, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1100.6796, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(750.7928, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(501.9669, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1322.3594, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1466.0103, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1115.9180, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(669.1398, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(500.3404, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(572.1453, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1079.4498, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1190.2981, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(961.9652, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(864.3528, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1161.0153, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1033.8870, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1165.4036, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1564.2249, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(804.7069, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1021.2001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(894.0481, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1182.4697, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(996.2073, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(930.9669, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(948.3333, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(891.6129, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1303.8000, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(588.0954, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1011.6417, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2413.6367, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1247.5347, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1153.8608, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.07367280606717226, 'recall': 0.09379310344827586, 'f1': 0.08252427184466019, 'number': 725}, 'P': {'precision': 0.09621993127147767, 'recall': 0.10579345088161209, 'f1': 0.10077984403119376, 'number': 794}, 'overall_precision': 0.08463251670378619, 'overall_recall': 0.10006583278472679, 'overall_f1': 0.09170437405731523, 'overall_accuracy': 0.6415919332530228}
			------------EPOCH 5---------------
Loss:  tensor(733.1559, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1191.4694, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1521.1823, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1503.2081, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1344.6298, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(945.4203, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(666.1895, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(866.9838, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(626.0381, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(409.7412, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1085.3860, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1325.7870, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(933.0336, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(531.0358, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(404.7965, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(466.9797, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(863.3267, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(946.8038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(832.1570, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(671.7646, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(851.0980, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(816.1808, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(951.6092, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1322.8479, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(662.6703, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(776.6652, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(650.3391, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(949.2200, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(827.1802, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(796.9648, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(843.8047, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(746.7312, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1140.6011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(455.6122, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(835.1276, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1998.4203, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(898.0315, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(973.3508, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.14166666666666666, 'recall': 0.2110344827586207, 'f1': 0.16952908587257617, 'number': 725}, 'P': {'precision': 0.1655359565807327, 'recall': 0.15365239294710328, 'f1': 0.15937295885042455, 'number': 794}, 'overall_precision': 0.15134837644468904, 'overall_recall': 0.18104015799868334, 'overall_f1': 0.16486810551558753, 'overall_accuracy': 0.6537500557712043}
			------------EPOCH 6---------------
Loss:  tensor(549.6930, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(964.3926, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1154.4167, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1149.2222, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1071.7119, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(732.9608, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(453.5159, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(643.0649, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(474.4784, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(321.6388, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(944.5925, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1119.7219, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(732.5181, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(396.2316, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(302.6008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(328.4089, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(737.1529, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(703.2623, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(732.3987, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(640.0935, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(693.1517, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(674.4551, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(777.4298, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1082.6138, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(576.5837, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(606.8312, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(488.9364, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(769.4396, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(717.0757, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(715.1064, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(799.3037, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(679.7032, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1100.5647, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(367.5211, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(720.4164, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1840.5159, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(686.2032, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(841.0907, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.17071197411003236, 'recall': 0.2910344827586207, 'f1': 0.21519632840387556, 'number': 725}, 'P': {'precision': 0.24336973478939158, 'recall': 0.1964735516372796, 'f1': 0.21742160278745645, 'number': 794}, 'overall_precision': 0.19552477357485348, 'overall_recall': 0.24160631994733378, 'overall_f1': 0.21613663133097763, 'overall_accuracy': 0.6511176549324053}
			------------EPOCH 7---------------
Loss:  tensor(397.4484, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(886.7881, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(965.1893, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(956.2906, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(913.1396, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(589.6783, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(347.4000, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(488.7703, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(403.7209, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(299.4583, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(838.7196, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(924.0162, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(653.3802, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(310.4417, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(288.4844, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(333.2591, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(645.0828, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(742.9910, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(795.8820, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(498.9208, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(592.9232, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(527.3228, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(763.4741, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(953.5588, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(437.4104, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(520.1425, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(499.4646, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(619.3098, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(598.3929, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(628.6345, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(755.8862, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(614.0645, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1138.5167, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(401.1070, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(651.3750, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1536.1567, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(609.3828, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(853.2753, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.20392156862745098, 'recall': 0.21517241379310345, 'f1': 0.20939597315436242, 'number': 725}, 'P': {'precision': 0.24311183144246354, 'recall': 0.3778337531486146, 'f1': 0.2958579881656805, 'number': 794}, 'overall_precision': 0.22811405702851426, 'overall_recall': 0.30019749835418036, 'overall_f1': 0.25923820352473, 'overall_accuracy': 0.6445143443537232}
			------------EPOCH 8---------------
Loss:  tensor(465.2894, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1080.7839, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1163.7295, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(938.9918, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1070.6631, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(591.4949, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(342.0284, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(551.3976, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(451.6361, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(383.2857, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(992.0045, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1059.2146, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(808.0929, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(311.9632, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(345.8482, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(400.5861, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(952.4746, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(886.8083, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(786.5256, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(703.7827, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(702.8024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(723.6380, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(639.9600, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1315.1304, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(506.6651, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(718.2011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(456.8695, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(783.3126, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(611.1848, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(613.0092, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(708.7637, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(629.0670, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1105.7328, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(468.5888, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(718.5714, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2125.2456, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(688.3052, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(817.6112, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.358974358974359, 'recall': 0.07724137931034483, 'f1': 0.1271282633371169, 'number': 725}, 'P': {'precision': 0.1919191919191919, 'recall': 0.3589420654911839, 'f1': 0.25010969723562965, 'number': 794}, 'overall_precision': 0.20780012187690433, 'overall_recall': 0.22448979591836735, 'overall_f1': 0.21582278481012657, 'overall_accuracy': 0.6214919912550751}
			------------EPOCH 9---------------
Loss:  tensor(519.1378, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(978.8426, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1362.2832, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1189.7433, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1142.9337, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(854.5828, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(516.0959, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(828.2185, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(549.5812, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(504.2769, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1207.7418, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1330.0186, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(610.9976, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(295.3126, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(194.1653, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(209.3100, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(697.7184, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(574.3818, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(545.8111, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(475.5771, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(596.5671, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(625.8965, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(648.8573, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(966.3411, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(541.7166, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(632.9097, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(564.7419, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(854.4973, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(663.1029, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(641.4397, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(616.0018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(509.1248, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1021.8607, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(410.0442, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(869.5339, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1615.5693, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(704.1903, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(847.3456, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.15625, 'recall': 0.2896551724137931, 'f1': 0.20299661672305463, 'number': 725}, 'P': {'precision': 0.11879699248120301, 'recall': 0.09949622166246852, 'f1': 0.10829335161069226, 'number': 794}, 'overall_precision': 0.14385266301642607, 'overall_recall': 0.1902567478604345, 'overall_f1': 0.16383219954648526, 'overall_accuracy': 0.6385133627805292}
			------------EPOCH 10---------------
Loss:  tensor(340.5266, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(643.4954, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1003.6722, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(874.0165, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(635.1650, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(509.0470, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(295.9274, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(497.5354, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(247.4872, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(217.0168, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(568.2736, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(560.1352, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(384.4574, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(198.7884, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(150.2268, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(155.1763, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(429.6942, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(453.0569, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(520.0292, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(353.0715, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(439.8315, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(444.7460, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(533.9933, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(826.8848, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(356.0715, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(403.2469, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(380.9161, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(444.8205, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(352.5891, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(386.4471, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(496.0301, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(377.7365, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(781.6787, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(263.0019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(358.2110, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(970.2662, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(359.0040, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(503.4205, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.20622568093385213, 'recall': 0.14620689655172414, 'f1': 0.17110573042776434, 'number': 725}, 'P': {'precision': 0.15301391035548687, 'recall': 0.24937027707808565, 'f1': 0.18965517241379312, 'number': 794}, 'overall_precision': 0.168141592920354, 'overall_recall': 0.20013166556945358, 'overall_f1': 0.1827472197174632, 'overall_accuracy': 0.6738723062508366}
			------------EPOCH 11---------------
Loss:  tensor(206.7605, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(454.5109, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(484.2545, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(457.3436, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(478.6731, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(320.7117, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(148.1336, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(282.4760, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(155.9766, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(177.8355, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(445.9121, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(437.1050, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(290.5198, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(143.8247, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(116.8547, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(119.9898, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(288.6366, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(321.3468, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(390.3751, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(249.7794, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(284.9523, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(349.7236, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(383.2306, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(556.6192, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(298.5558, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(338.5606, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(262.8159, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(400.6260, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(226.1802, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(299.5520, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(365.9974, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(249.2416, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(604.1398, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(185.1887, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(289.3574, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(711.6337, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(242.7953, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(413.3514, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.15733590733590733, 'recall': 0.22482758620689655, 'f1': 0.18512208972174898, 'number': 725}, 'P': {'precision': 0.16233766233766234, 'recall': 0.22040302267002518, 'f1': 0.18696581196581194, 'number': 794}, 'overall_precision': 0.1598864711447493, 'overall_recall': 0.22251481237656354, 'overall_f1': 0.18607211670795484, 'overall_accuracy': 0.6777762905456655}
			------------EPOCH 12---------------
Loss:  tensor(144.3579, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(345.1618, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(312.0251, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(356.4738, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(357.6278, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(220.3293, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(107.9804, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(223.1996, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(109.9495, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(156.7364, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(321.9855, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(310.2380, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(213.8790, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(107.4929, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(90.4869, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(90.1468, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(205.0623, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(234.4166, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(293.3079, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(168.0500, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(177.0064, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(244.8599, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(305.1109, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(477.9875, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(240.3472, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(268.4113, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(212.5927, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(237.4338, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(145.6052, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(197.5885, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(258.8723, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(183.8566, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(478.2144, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(135.3221, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(173.9598, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(543.7701, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(192.9781, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(302.5297, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.17032967032967034, 'recall': 0.17103448275862068, 'f1': 0.17068134893324155, 'number': 725}, 'P': {'precision': 0.2029780564263323, 'recall': 0.3261964735516373, 'f1': 0.2502415458937198, 'number': 794}, 'overall_precision': 0.19111776447105788, 'overall_recall': 0.2521395655036208, 'overall_f1': 0.21742832812943516, 'overall_accuracy': 0.6806540846829965}
			------------EPOCH 13---------------
Loss:  tensor(103.7888, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(277.7068, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(191.5916, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(223.3611, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(269.9989, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(169.1985, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(69.3956, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(169.5488, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(77.4099, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(136.7384, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(248.6673, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(230.5726, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(163.6865, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(77.8099, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(72.8195, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(61.5006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(155.0075, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(179.6356, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(206.4125, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(127.9691, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(129.8162, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(190.0724, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(225.0695, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(363.6828, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(188.5157, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(193.4385, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(167.8910, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(180.8362, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(101.5141, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(148.5341, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(178.8663, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(134.4115, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(357.8627, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(104.7594, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(126.3965, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(385.7086, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(134.1073, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(220.4008, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.1687041564792176, 'recall': 0.19034482758620688, 'f1': 0.17887232663642252, 'number': 725}, 'P': {'precision': 0.1825525040387722, 'recall': 0.28463476070528965, 'f1': 0.22244094488188976, 'number': 794}, 'overall_precision': 0.17704280155642024, 'overall_recall': 0.23963133640552994, 'overall_f1': 0.20363636363636364, 'overall_accuracy': 0.678155534734306}
			------------EPOCH 14---------------
Loss:  tensor(81.4113, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(202.9576, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(130.4244, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(179.1533, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(194.1742, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(124.4582, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.7053, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(144.7475, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.5818, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(125.0712, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(197.8586, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(167.0381, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(118.3410, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.3509, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.1241, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.3158, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(110.0014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(133.4879, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(151.9031, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(95.0020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(83.9818, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(134.3863, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(171.7042, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(307.5834, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(153.2888, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(150.3686, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(123.2241, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(129.6066, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(76.6151, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(111.6407, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(120.9399, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(108.6298, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(272.8067, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(75.5673, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(92.2343, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(302.0299, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(92.8526, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(157.3622, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.16926770708283315, 'recall': 0.19448275862068964, 'f1': 0.18100128369704752, 'number': 725}, 'P': {'precision': 0.1928, 'recall': 0.3035264483627204, 'f1': 0.23581213307240706, 'number': 794}, 'overall_precision': 0.18338934229476717, 'overall_recall': 0.2514812376563529, 'overall_f1': 0.21210438645197113, 'overall_accuracy': 0.6759916120108865}
			------------EPOCH 15---------------
Loss:  tensor(60.2723, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(155.0952, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(91.8710, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(143.3885, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(151.8916, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(94.3095, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.4765, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(118.7809, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.3001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(101.9481, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(148.0883, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(124.6758, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(92.8315, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.0122, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.1761, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.0291, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(79.0279, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(110.0963, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(107.0048, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(88.6671, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(78.4443, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(108.4298, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(125.7830, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(258.4625, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(126.5049, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(142.9061, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(96.1705, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(110.0630, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.7160, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(81.5517, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(81.8056, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(88.3012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(211.1261, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.0720, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(76.5647, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(248.9312, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(66.0143, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(106.5969, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.1813186813186813, 'recall': 0.1820689655172414, 'f1': 0.18169304886441842, 'number': 725}, 'P': {'precision': 0.20030816640986132, 'recall': 0.327455919395466, 'f1': 0.24856596558317398, 'number': 794}, 'overall_precision': 0.1934846989141165, 'overall_recall': 0.25806451612903225, 'overall_f1': 0.22115655853314528, 'overall_accuracy': 0.675790835675724}
			------------EPOCH 16---------------
Loss:  tensor(43.2002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(131.1272, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(67.1630, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(133.5135, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(108.7901, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(75.9078, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.7482, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(94.5195, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.4597, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(64.4757, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(118.9573, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(96.5917, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(67.1553, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.7651, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.5179, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.9151, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(59.6756, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(93.0259, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(82.2745, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(107.2765, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(123.0741, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(113.2832, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(93.4995, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(255.4219, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(84.3474, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(81.0876, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(71.7969, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(82.5671, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.6250, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(59.0977, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(65.6641, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(79.4829, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(169.3689, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.5469, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.1733, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(271.2000, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(68.3264, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(80.2078, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.1919191919191919, 'recall': 0.23586206896551723, 'f1': 0.2116336633663366, 'number': 725}, 'P': {'precision': 0.2012779552715655, 'recall': 0.31738035264483627, 'f1': 0.2463343108504399, 'number': 794}, 'overall_precision': 0.19738684087727484, 'overall_recall': 0.2784726793943384, 'overall_f1': 0.2310212998361551, 'overall_accuracy': 0.679003257038326}
			------------EPOCH 17---------------
Loss:  tensor(31.7930, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(89.1600, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.5488, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(134.8883, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(84.1360, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.4644, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.5917, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(82.3197, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.3481, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.4167, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(122.8066, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(96.5831, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.4363, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.2380, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.3137, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.0557, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(65.4729, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(97.6291, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.0081, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(72.5003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(85.7848, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(75.1446, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(77.7774, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(206.7310, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(79.4611, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(83.7628, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(67.5775, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(73.4570, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.7691, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.1416, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.8505, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(59.3718, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(144.2749, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.0962, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.3238, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(255.6575, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(67.1965, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(77.3791, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.16732026143790849, 'recall': 0.17655172413793102, 'f1': 0.17181208053691271, 'number': 725}, 'P': {'precision': 0.11507052709725316, 'recall': 0.1952141057934509, 'f1': 0.1447921531994395, 'number': 794}, 'overall_precision': 0.13399621212121213, 'overall_recall': 0.18630678077682686, 'overall_f1': 0.15587992288625724, 'overall_accuracy': 0.6806540846829965}
			------------EPOCH 18---------------
Loss:  tensor(30.9805, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(68.1061, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(73.8616, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(115.8328, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(79.1865, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(59.9688, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.3261, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(81.7122, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.6333, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.6298, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(78.8167, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(62.0051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.3317, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.7836, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.2301, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.2525, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.4776, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(70.5576, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.8819, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.9885, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(61.0972, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.5784, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.1373, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(182.8490, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(84.5952, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(89.4982, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.7600, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(75.8050, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.2779, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.4695, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(62.6983, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(68.6595, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(152.5049, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.9043, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.8107, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(255.3996, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.8820, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.5311, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.16201859229747675, 'recall': 0.16827586206896553, 'f1': 0.1650879566982409, 'number': 725}, 'P': {'precision': 0.1986706056129985, 'recall': 0.33879093198992444, 'f1': 0.2504655493482309, 'number': 794}, 'overall_precision': 0.1855719031798766, 'overall_recall': 0.2574061882817643, 'overall_f1': 0.21566464423607282, 'overall_accuracy': 0.6746307946281176}
			------------EPOCH 19---------------
Loss:  tensor(20.9750, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.0759, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.4897, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(74.0192, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.7404, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.0677, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.4600, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(74.9705, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.0558, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.9907, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(75.9379, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(74.2675, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.2731, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.6682, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.5717, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.0601, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(82.3555, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(107.3036, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(66.2629, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.2203, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.3588, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.0890, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(83.0281, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(232.4135, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(67.8352, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(90.5509, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.6389, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.0413, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.7662, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.1161, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.4503, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.2401, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(95.6041, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.8315, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.5164, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(196.7470, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.1996, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.6876, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.20757020757020758, 'recall': 0.23448275862068965, 'f1': 0.22020725388601037, 'number': 725}, 'P': {'precision': 0.2129032258064516, 'recall': 0.33249370277078083, 'f1': 0.25958702064896755, 'number': 794}, 'overall_precision': 0.21078193297717338, 'overall_recall': 0.2857142857142857, 'overall_f1': 0.242593627724986, 'overall_accuracy': 0.6682728773479677}
			------------EPOCH 20---------------
Loss:  tensor(32.4460, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(89.5424, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.3259, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(91.1790, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(101.9930, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(87.6197, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.0379, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(69.7243, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.5049, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.5075, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(87.1354, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(77.2677, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.1098, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.6336, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.1556, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.8438, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.3341, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(85.8804, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.3623, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.4693, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.6548, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.7128, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.0827, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(153.8495, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.6136, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(75.0137, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.3703, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.3217, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.0954, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.3514, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.5067, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.1967, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(123.3860, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.9026, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.8469, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(185.5942, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.7915, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.9331, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.18309859154929578, 'recall': 0.19724137931034483, 'f1': 0.1899070385126162, 'number': 725}, 'P': {'precision': 0.13609467455621302, 'recall': 0.23173803526448364, 'f1': 0.17148182665424044, 'number': 794}, 'overall_precision': 0.15330520393811534, 'overall_recall': 0.2152732060566162, 'overall_f1': 0.17907995618838993, 'overall_accuracy': 0.6833087940034801}
			------------EPOCH 21---------------
Loss:  tensor(23.0097, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.3884, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.4342, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(88.6028, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.4168, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.7502, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.8462, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(122.5624, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.2415, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.0631, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.8821, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.3832, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.4080, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.2857, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.2001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.4809, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.7976, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(62.5301, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.2429, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.8976, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.0757, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.9880, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.8330, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(127.6993, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(75.0684, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(110.9555, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.6330, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(75.1809, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.8295, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.0297, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.8060, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.0814, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(82.9639, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.5835, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.4722, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(189.0674, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.5412, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.0598, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.20474406991260924, 'recall': 0.22620689655172413, 'f1': 0.21494102228047182, 'number': 725}, 'P': {'precision': 0.2032967032967033, 'recall': 0.3261964735516373, 'f1': 0.2504835589941973, 'number': 794}, 'overall_precision': 0.203855421686747, 'overall_recall': 0.2784726793943384, 'overall_f1': 0.23539232053422374, 'overall_accuracy': 0.6864765984027127}
			------------EPOCH 22---------------
Loss:  tensor(10.4083, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.7063, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.0596, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.4718, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.0738, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.7234, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.8260, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(108.6826, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.4100, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.2731, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.9624, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.1957, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.9497, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.0993, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.8848, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.2148, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(79.3726, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(66.1731, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.9153, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.3109, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.9049, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.6312, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.0307, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(142.0842, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.6555, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(93.3161, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.8283, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(81.8268, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.0406, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.1906, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.4369, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.3194, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(71.2583, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.7203, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.7125, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(177.1879, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.8725, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.1106, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.1896774193548387, 'recall': 0.20275862068965517, 'f1': 0.196, 'number': 725}, 'P': {'precision': 0.19047619047619047, 'recall': 0.30730478589420657, 'f1': 0.23518072289156627, 'number': 794}, 'overall_precision': 0.19017509727626458, 'overall_recall': 0.2574061882817643, 'overall_f1': 0.21874125874125872, 'overall_accuracy': 0.6728684245750234}
			------------EPOCH 23---------------
Loss:  tensor(8.9990, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.5942, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.4820, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.9597, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.4753, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.3862, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.9083, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(102.0493, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.4155, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.2874, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.6167, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.1966, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.9407, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.9226, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.4722, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.8797, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.8192, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(73.3829, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.4578, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.2925, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.8509, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.9242, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.1955, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(135.5446, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.3020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.9649, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(68.5941, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.9683, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.0989, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.1043, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.1590, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.9979, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.2758, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.6975, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.2870, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(132.6580, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.7512, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.6661, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.17391304347826086, 'recall': 0.19310344827586207, 'f1': 0.1830065359477124, 'number': 725}, 'P': {'precision': 0.1493411420204978, 'recall': 0.25692695214105793, 'f1': 0.18888888888888888, 'number': 794}, 'overall_precision': 0.15845232611699678, 'overall_recall': 0.22646477946017116, 'overall_f1': 0.18644986449864498, 'overall_accuracy': 0.674251550439477}
			------------EPOCH 24---------------
Loss:  tensor(7.5221, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.1021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.2466, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.6060, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.8754, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.6912, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.3007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(81.7250, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.6822, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.2733, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.5104, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.2786, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.5330, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.8990, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.3903, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.2134, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.0974, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(62.7650, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.2111, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.4377, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.9163, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.1920, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.1536, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(88.7829, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.0594, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.1259, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.6345, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.1542, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.7473, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.5344, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.3220, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.0227, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.6476, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.8333, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.7131, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(106.0809, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.1393, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.2799, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.19131614654002713, 'recall': 0.19448275862068964, 'f1': 0.1928864569083447, 'number': 725}, 'P': {'precision': 0.2173202614379085, 'recall': 0.3350125944584383, 'f1': 0.2636273538156591, 'number': 794}, 'overall_precision': 0.20754716981132076, 'overall_recall': 0.26793943383805136, 'overall_f1': 0.2339080459770115, 'overall_accuracy': 0.6762593137911034}
			------------EPOCH 25---------------
Loss:  tensor(6.2309, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.3099, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.5071, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.2743, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.6979, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.1365, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.8691, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(66.1442, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.5729, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.1509, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.9376, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.6527, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.0959, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.9263, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.6301, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.1371, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.8066, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.8769, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.3053, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.4188, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.6400, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.6813, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.2672, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.3130, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.2015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.9271, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.7233, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.4137, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.7450, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.0800, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.4844, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.5327, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.0072, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.0189, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.0762, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(69.8821, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.0341, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.9364, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.1706102117061021, 'recall': 0.1889655172413793, 'f1': 0.1793193717277487, 'number': 725}, 'P': {'precision': 0.18615384615384614, 'recall': 0.3047858942065491, 'f1': 0.23113658070678128, 'number': 794}, 'overall_precision': 0.1802187351402758, 'overall_recall': 0.24950625411454905, 'overall_f1': 0.20927664273881838, 'overall_accuracy': 0.6732030518002945}
			------------EPOCH 26---------------
Loss:  tensor(5.2660, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.3221, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.2329, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.0127, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.6490, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.1292, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.3280, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.0105, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.6652, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.0122, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.4800, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.1458, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.1162, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.7307, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.7334, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.1603, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.3590, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.1947, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.5931, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.0408, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.5738, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.4301, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.9605, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.2767, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.0429, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.8811, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.9059, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.7825, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.9952, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.1446, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.7453, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.0419, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.5027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.6676, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.0075, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.4910, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.9621, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.5959, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.17596034696406443, 'recall': 0.19586206896551725, 'f1': 0.185378590078329, 'number': 725}, 'P': {'precision': 0.18838304552590268, 'recall': 0.3022670025188917, 'f1': 0.2321083172147002, 'number': 794}, 'overall_precision': 0.18356559346468043, 'overall_recall': 0.2514812376563529, 'overall_f1': 0.21222222222222223, 'overall_accuracy': 0.6768393343149065}
			------------EPOCH 27---------------
Loss:  tensor(4.4415, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.1963, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.8252, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.6793, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.3395, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.6580, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.9353, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.8331, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2131, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.3468, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(118.4288, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(62.1611, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.5840, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.1982, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.7991, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.0281, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.7620, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.3007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.5012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.9687, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.6707, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.2551, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.5744, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.0924, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.0441, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.0384, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.7402, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.7274, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5293, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.1952, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.4504, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.0558, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.2798, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2109, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.4220, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.4590, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.1282, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.1838, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.19047619047619047, 'recall': 0.20413793103448277, 'f1': 0.19707057256990682, 'number': 725}, 'P': {'precision': 0.19809825673534073, 'recall': 0.3148614609571788, 'f1': 0.2431906614785992, 'number': 794}, 'overall_precision': 0.19519372241294752, 'overall_recall': 0.26201448321263987, 'overall_f1': 0.2237211916807195, 'overall_accuracy': 0.675790835675724}
			------------EPOCH 28---------------
Loss:  tensor(3.8127, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.8226, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.8331, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.0935, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.3865, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.3626, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6919, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.5726, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.8579, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.9460, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.3346, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.2822, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.0262, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.8565, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5428, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.7337, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.9744, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.7316, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.7487, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.2809, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.3045, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.7269, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.8162, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.7452, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.4908, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.1534, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.8577, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.9460, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2125, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.6120, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.8728, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.8352, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.8768, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.9089, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.8251, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.0481, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.7237, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.7912, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.19096774193548388, 'recall': 0.20413793103448277, 'f1': 0.19733333333333333, 'number': 725}, 'P': {'precision': 0.1985981308411215, 'recall': 0.3211586901763224, 'f1': 0.24542829643888356, 'number': 794}, 'overall_precision': 0.195726080621661, 'overall_recall': 0.2653061224489796, 'overall_f1': 0.2252655114589156, 'overall_accuracy': 0.6796278945254985}
			------------EPOCH 29---------------
Loss:  tensor(3.3590, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.2987, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.1766, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.4559, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.9901, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.5065, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4987, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.3478, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5991, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6373, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.6283, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.9473, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.7525, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6657, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3502, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5135, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.3711, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.6919, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.3248, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.6158, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.0467, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.3837, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.3616, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.7508, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.0618, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.3237, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.6525, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.7899, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.0313, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.2005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.1140, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.0453, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.9032, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.7596, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.2507, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.9492, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.7500, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.9953, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.19041450777202074, 'recall': 0.20275862068965517, 'f1': 0.1963927855711423, 'number': 725}, 'P': {'precision': 0.19844357976653695, 'recall': 0.3211586901763224, 'f1': 0.2453102453102453, 'number': 794}, 'overall_precision': 0.19543023821098687, 'overall_recall': 0.2646477946017117, 'overall_f1': 0.22483221476510068, 'overall_accuracy': 0.6766831749431134}
			------------EPOCH 30---------------
Loss:  tensor(2.9107, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.8587, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.6376, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(59.5511, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.6774, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.3722, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4509, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.8484, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5393, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5195, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.0894, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.1385, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.4803, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3308, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1846, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.1236, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(77.5743, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.0150, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.1988, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.2191, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.4601, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.0754, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.4136, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.5227, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.1126, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.1050, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.5315, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.9903, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.8234, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.0610, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.3703, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.8147, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6507, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.9849, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(99.1829, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.2934, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.0370, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.17633410672853828, 'recall': 0.2096551724137931, 'f1': 0.19155639571518587, 'number': 725}, 'P': {'precision': 0.1920260374288039, 'recall': 0.2972292191435768, 'f1': 0.2333168561542264, 'number': 794}, 'overall_precision': 0.18555714968914394, 'overall_recall': 0.2554312047399605, 'overall_f1': 0.2149584487534626, 'overall_accuracy': 0.6762816222727881}


		-------------RUN 5-----------
			------------EPOCH 1---------------
Loss:  tensor(2802.9231, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2950.0581, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3126.7686, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2188.7476, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1743.7378, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1895.2872, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2588.0747, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1163.9222, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3007.8901, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3054.2583, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4221.3379, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2715.0112, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2552.8687, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1727.7886, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1882.9198, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1275.3962, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1643.3594, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2372.7717, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2332., device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2263.8960, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1593.5272, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3031.6499, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1225.7764, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1459.4133, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2147.1003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2124.5156, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1237.0826, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1537.8313, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2120.5298, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1895.4902, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1728.4528, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1791.6152, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1667.2441, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1047.3633, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1633.9031, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1299.1914, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.018604651162790697, 'recall': 0.011283497884344146, 'f1': 0.014047410008779631, 'number': 709}, 'P': {'precision': 0.05119735755573906, 'recall': 0.08147174770039421, 'f1': 0.06288032454361055, 'number': 761}, 'overall_precision': 0.042656916514320534, 'overall_recall': 0.047619047619047616, 'overall_f1': 0.04500160720025715, 'overall_accuracy': 0.571842099620473}
			------------EPOCH 2---------------
Loss:  tensor(1679.3683, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2002.7891, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2312.6477, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1440.6350, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1322.0471, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1450.3652, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1979.9766, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(769.8162, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2239.4612, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2138.5249, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3495.1021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2118.0952, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2052.4409, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1331.9132, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1574.8860, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1015.2692, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1317.8899, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2162.9502, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1739.1882, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1706.0039, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1329.2078, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2651.5461, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1024.9238, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1317.7354, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1815.6782, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1909.6970, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1070.0497, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1249.7643, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1818.3604, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1529.9496, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1658.9072, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1460.8423, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1443.0885, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(922.9896, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1452.6848, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1196.9663, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.03761755485893417, 'recall': 0.03385049365303244, 'f1': 0.0356347438752784, 'number': 709}, 'P': {'precision': 0.06413124533929904, 'recall': 0.11300919842312747, 'f1': 0.08182683158896291, 'number': 761}, 'overall_precision': 0.05558362809499747, 'overall_recall': 0.07482993197278912, 'overall_f1': 0.06378660481298927, 'overall_accuracy': 0.5995025409010013}
			------------EPOCH 3---------------
Loss:  tensor(1475.4441, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1818.0645, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2131.1206, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1251.2458, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1177.5310, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1295.8148, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1803.7202, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(646.6781, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1927.0654, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1849.1949, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3348.4907, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1931.1724, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1773.2074, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1191.7083, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1323.6589, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(936.0632, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1204.7500, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2012.6553, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1421.3674, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1445.9445, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1153.0845, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2346.3921, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(885.9012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1116.8718, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1462.0103, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1675.5737, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1013.7415, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1139.4180, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1566.9519, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1247.7439, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1436.9430, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1286.9980, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1291.0131, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(812.5454, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1258.6270, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1093.1042, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.0914826498422713, 'recall': 0.1227080394922426, 'f1': 0.10481927710843374, 'number': 709}, 'P': {'precision': 0.1613198900091659, 'recall': 0.23127463863337713, 'f1': 0.19006479481641467, 'number': 761}, 'overall_precision': 0.1287952987267385, 'overall_recall': 0.17891156462585034, 'overall_f1': 0.1497722095671982, 'overall_accuracy': 0.6137401633895834}
			------------EPOCH 4---------------
Loss:  tensor(1368.0741, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1738.1891, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2091.1355, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1053.9238, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1021.6735, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1102.9432, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1665.9944, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(498.4346, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1645.1926, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1690.5898, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3002.6047, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1707.6539, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1468.0715, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1042.0530, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1133.2777, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(846.7039, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1131.3763, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1803.1710, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1232.8611, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1230.1105, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1027.7842, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2230.0540, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(784.5603, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(965.8992, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1274.6298, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1467.1118, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(980.2653, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1069.6676, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1384.2144, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(991.8055, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1136.6138, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1148.9860, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1104.7710, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(694.3953, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1024.1377, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(923.0508, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.12418906394810009, 'recall': 0.18899858956276447, 'f1': 0.1498881431767338, 'number': 709}, 'P': {'precision': 0.22580645161290322, 'recall': 0.3403416557161629, 'f1': 0.27148846960167716, 'number': 761}, 'overall_precision': 0.1765498652291105, 'overall_recall': 0.2673469387755102, 'overall_f1': 0.21266233766233766, 'overall_accuracy': 0.6123464202242854}
			------------EPOCH 5---------------
Loss:  tensor(1160.0057, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1510.6221, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1902.9468, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(928.9482, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(978.4905, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1032.9458, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1500.2175, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(415.7642, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1486.6653, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1538.1042, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2738.6450, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1523.0974, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1476.0852, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(965.7523, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1169.5377, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(731.8073, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(859.8630, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1495.2507, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(956.8107, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1060.1407, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(815.9128, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1767.4783, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(644.0037, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(816.3745, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(935.4330, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1201.5476, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(850.3306, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(878.6025, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1281.2975, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(884.0256, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(926.4296, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1018.4897, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(977.9946, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(665.2328, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(906.8465, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(763.2382, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.0975800156128025, 'recall': 0.1763046544428773, 'f1': 0.1256281407035176, 'number': 709}, 'P': {'precision': 0.02929427430093209, 'recall': 0.02890932982917214, 'f1': 0.029100529100529102, 'number': 761}, 'overall_precision': 0.07234251968503937, 'overall_recall': 0.1, 'overall_f1': 0.0839520274129069, 'overall_accuracy': 0.6270986555739005}
			------------EPOCH 6---------------
Loss:  tensor(1071.1917, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1173.4603, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1429.7861, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(865.9865, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(809.7125, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(827.3823, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1385.2189, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(435.4981, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1280.9398, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1367.7343, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1935.5447, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1034.3663, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1041.7782, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(738.3081, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(907.0582, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(629.0601, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(684.0751, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1229.5503, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(749.6200, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(896.7023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(750.7981, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1614.3456, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(660.3569, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(781.9554, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(790.5121, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1243.2942, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(779.6713, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(693.9261, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(915.2897, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(677.9108, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(814.9150, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(790.3668, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(693.2841, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(501.5535, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(763.4932, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(598.8853, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.10186199342825848, 'recall': 0.2623413258110014, 'f1': 0.1467455621301775, 'number': 709}, 'P': {'precision': 0.0297029702970297, 'recall': 0.01971090670170828, 'f1': 0.023696682464454978, 'number': 761}, 'overall_precision': 0.08622908622908623, 'overall_recall': 0.13673469387755102, 'overall_f1': 0.10576164167324388, 'overall_accuracy': 0.5951283315822201}
			------------EPOCH 7---------------
Loss:  tensor(863.6782, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(968.3417, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1269.6549, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(687.3785, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(675.3981, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(709.1119, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1225.5796, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(356.4557, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1139.4823, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1348.4504, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1757.5565, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(919.9694, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1001.1085, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(746.1635, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(785.8380, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(638.2874, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(557.6195, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1172.4082, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(654.8042, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(681.1405, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(517.4050, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1259.0610, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(484.3027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(531.0764, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(660.5507, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1007.3510, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(655.1525, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(588.7642, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(801.1877, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(568.0114, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(588.7238, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(762.7103, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(652.4781, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(391.7549, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(600.4393, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(440.6051, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.17724867724867724, 'recall': 0.37799717912552894, 'f1': 0.2413327330031517, 'number': 709}, 'P': {'precision': 0.2736220472440945, 'recall': 0.18265440210249673, 'f1': 0.219070133963751, 'number': 761}, 'overall_precision': 0.2014851485148515, 'overall_recall': 0.27687074829931974, 'overall_f1': 0.23323782234957022, 'overall_accuracy': 0.5964148637348028}
			------------EPOCH 8---------------
Loss:  tensor(774.4189, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1040.1389, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1285.9980, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(636.3541, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(741.3377, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(644.3198, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1135.7898, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(277.1095, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1221.7708, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1255.0273, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1903.9788, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(883.0716, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(764.9638, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(569.9902, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(776.7433, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(527.9161, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(486.6104, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(989.4822, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(473.4329, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(501.3256, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(395.7943, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1086.9365, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(325.5715, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(381.8600, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(498.8113, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(740.2283, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(531.3007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(463.1293, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(769.1453, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(531.3002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(482.0331, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(683.4716, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(691.7212, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(477.3972, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(892.6724, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(417.0787, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.15350877192982457, 'recall': 0.09873060648801128, 'f1': 0.12017167381974249, 'number': 709}, 'P': {'precision': 0.08166576527852894, 'recall': 0.19842312746386334, 'f1': 0.11570881226053639, 'number': 761}, 'overall_precision': 0.09587852494577007, 'overall_recall': 0.15034013605442176, 'overall_f1': 0.1170860927152318, 'overall_accuracy': 0.6368334155284431}
			------------EPOCH 9---------------
Loss:  tensor(924.7029, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1261.7004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1313.4717, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(683.4200, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(757.3268, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(810.9263, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1341.5646, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(276.2043, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(758.0811, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(938.2930, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1353.7235, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(640.0808, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(549.6655, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(422.8085, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(511.3297, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(441.1950, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(526.6542, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(985.3436, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(561.4869, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(528.9262, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(506.4758, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1259.7708, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(383.6646, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(583.6685, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(618.5666, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1041.8400, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(660.3547, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(660.6958, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(767.5827, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(455.8503, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(452.8104, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(683.9432, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(457.9470, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(287.8140, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(358.6910, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(312.0927, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.14597815292949354, 'recall': 0.2073342736248237, 'f1': 0.1713286713286713, 'number': 709}, 'P': {'precision': 0.17296389588581024, 'recall': 0.2706964520367937, 'f1': 0.21106557377049182, 'number': 761}, 'overall_precision': 0.16060054595086443, 'overall_recall': 0.24013605442176872, 'overall_f1': 0.1924754634678299, 'overall_accuracy': 0.6599266676673028}
			------------EPOCH 10---------------
Loss:  tensor(574.4693, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(663.1870, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(685.4758, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(386.3054, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(555.8022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(481.6313, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1015.6104, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(164.7472, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(629.9849, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(853.1016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1440.1213, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(580.4247, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(606.5282, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(392.7596, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(519.3550, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(468.9119, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(350.6611, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(686.9672, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(401.7824, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(371.1311, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(256.4434, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(717.3410, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(225.0374, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(252.5778, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(292.5753, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(542.1000, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(335.9804, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(225.4940, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(442.2575, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(310.4473, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(279.5757, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(354.8936, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(326.2581, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(227.6977, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(286.2018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(258.6126, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.11333794056668971, 'recall': 0.231311706629055, 'f1': 0.1521335807050093, 'number': 709}, 'P': {'precision': 0.07754010695187166, 'recall': 0.07621550591327202, 'f1': 0.07687210072895959, 'number': 761}, 'overall_precision': 0.10113895216400912, 'overall_recall': 0.1510204081632653, 'overall_f1': 0.12114597544338336, 'overall_accuracy': 0.6447027038617407}
			------------EPOCH 11---------------
Loss:  tensor(523.6910, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(509.8193, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(595.2299, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(302.2167, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(369.3156, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(342.7321, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(766.5552, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(162.2694, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(428.4022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(743.0796, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1073.8038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(555.0785, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(380.3243, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(262.2318, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(376.0762, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(340.3394, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(245.0328, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(561.4513, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(330.2805, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(322.6300, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(242.1573, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(611.6913, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(218.3721, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(241.2888, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(241.7291, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(563.5686, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(246.6371, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(232.9459, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(450.5258, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(267.4961, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(261.6116, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(344.8414, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(282.2331, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(164.0525, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(234.7808, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(197.3710, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.16103896103896104, 'recall': 0.17489421720733428, 'f1': 0.16768086544962812, 'number': 709}, 'P': {'precision': 0.11150317572335922, 'recall': 0.2076215505913272, 'f1': 0.1450872359963269, 'number': 761}, 'overall_precision': 0.1289437585733882, 'overall_recall': 0.19183673469387755, 'overall_f1': 0.1542247744052502, 'overall_accuracy': 0.6664879816454746}
			------------EPOCH 12---------------
Loss:  tensor(346.8877, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(415.8591, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(442.4113, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(209.1041, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(309.1073, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(261.9703, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(647.5492, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(99.6063, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(320.6356, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(550.7413, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(818.7068, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(391.3469, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(295.4737, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(186.6779, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(266.7331, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(314.5562, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(220.7863, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(478.4846, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(250.9011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(253.8259, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(165.4397, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(437.3475, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(130.2392, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(161.0694, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(168.0190, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(366.0668, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(191.8881, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(129.4009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(296.3299, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(189.4459, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(270.7565, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(202.3275, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(180.2472, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(130.8215, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(171.1424, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(154.6684, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.15955882352941175, 'recall': 0.306064880112835, 'f1': 0.2097631706138231, 'number': 709}, 'P': {'precision': 0.16454352441613587, 'recall': 0.20367936925098554, 'f1': 0.18203170874926597, 'number': 761}, 'overall_precision': 0.16159860990443092, 'overall_recall': 0.2530612244897959, 'overall_f1': 0.1972428419936373, 'overall_accuracy': 0.6575036987799386}
			------------EPOCH 13---------------
Loss:  tensor(299.1461, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(351.9950, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(348.6636, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(183.2547, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(223.8110, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(158.5765, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(512.0362, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(71.3661, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(217.5072, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(442.3073, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(628.1195, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(248.3946, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(210.4292, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(120.4052, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(179.9555, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(241.9647, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(159.2209, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(379.1725, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(154.1462, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(154.4903, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(127.2424, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(325.5595, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(103.4286, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(109.2362, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(134.3477, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(263.9920, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(120.8018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(97.7706, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(226.7990, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(120.9449, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(148.6746, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(114.8225, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(124.1574, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(105.9946, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(129.5929, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(112.1769, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.17543859649122806, 'recall': 0.18335684062059238, 'f1': 0.1793103448275862, 'number': 709}, 'P': {'precision': 0.11119030648610122, 'recall': 0.2049934296977661, 'f1': 0.14417744916820702, 'number': 761}, 'overall_precision': 0.1333955223880597, 'overall_recall': 0.1945578231292517, 'overall_f1': 0.15827338129496404, 'overall_accuracy': 0.6736711194973948}
			------------EPOCH 14---------------
Loss:  tensor(230.4888, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(273.8340, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(262.9867, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(123.1020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(170.2889, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(142.6664, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(414.7266, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.0316, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(168.9745, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(317.2237, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(475.4532, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(202.3659, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(133.5420, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(72.5671, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(129.6475, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(192.9494, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(117.2631, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(301.0915, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(119.9076, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(128.1942, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(107.6659, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(242.1510, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(82.0008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(89.4516, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(129.4411, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(238.7588, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(112.0755, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(75.2504, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(179.5191, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(93.0723, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(133.8377, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(85.6049, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(68.9883, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(69.1401, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(102.9240, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(86.0791, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.14216867469879518, 'recall': 0.24964739069111425, 'f1': 0.1811668372569089, 'number': 709}, 'P': {'precision': 0.11826923076923077, 'recall': 0.1616294349540079, 'f1': 0.1365907828983898, 'number': 761}, 'overall_precision': 0.13129102844638948, 'overall_recall': 0.20408163265306123, 'overall_f1': 0.15978695073235685, 'overall_accuracy': 0.6724489139524412}
			------------EPOCH 15---------------
Loss:  tensor(166.8433, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(202.8549, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(207.5872, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(92.7496, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(113.8530, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(93.7272, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(296.6151, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.8112, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(136.5370, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(259.8028, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(362.5534, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(142.6348, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(102.1862, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.1047, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(81.3128, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(150.2289, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(96.7968, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(281.6963, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(82.5875, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(96.6045, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(88.4720, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(206.2301, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(67.3168, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(61.8860, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(71.1441, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(139.5567, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(67.4959, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.6665, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(111.3096, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(68.5883, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(93.6103, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.0031, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.2053, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(62.0523, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(69.2816, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.4852, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.1655011655011655, 'recall': 0.2002820874471086, 'f1': 0.18123803446075304, 'number': 709}, 'P': {'precision': 0.15326821938392185, 'recall': 0.2680683311432326, 'f1': 0.19502868068833648, 'number': 761}, 'overall_precision': 0.15806304248515304, 'overall_recall': 0.23537414965986395, 'overall_f1': 0.1891227111232577, 'overall_accuracy': 0.6740570791431696}
			------------EPOCH 16---------------
Loss:  tensor(124.7686, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(167.9056, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(161.7066, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(71.3103, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(78.4916, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(66.0592, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(220.2455, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.3076, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(114.5314, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(214.3066, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(296.8734, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(95.9002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(74.7003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.9521, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.2349, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(111.2866, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(61.3441, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(267.9349, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(114.5921, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(94.7004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.8947, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(170.9427, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.6240, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.6056, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.9829, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(114.0124, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.9912, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.3393, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(89.4079, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.0407, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(88.3395, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.4679, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.8846, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.2224, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.4954, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.6059, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.16034985422740525, 'recall': 0.23272214386459802, 'f1': 0.189873417721519, 'number': 709}, 'P': {'precision': 0.13459950454170108, 'recall': 0.21419185282522996, 'f1': 0.16531440162271807, 'number': 761}, 'overall_precision': 0.14642857142857144, 'overall_recall': 0.22312925170068026, 'overall_f1': 0.17681940700808627, 'overall_accuracy': 0.6803396444882819}
			------------EPOCH 17---------------
Loss:  tensor(91.3657, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(156.7986, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(147.5804, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.1616, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(59.4380, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.4936, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(172.7339, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.4985, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(103.0366, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(208.2576, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(317.4992, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(87.9170, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(89.3174, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.6132, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.8718, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(89.3850, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.9195, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(248.6666, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(165.2027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(108.8973, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.1518, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(136.6584, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.0076, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.7131, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.3331, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(105.7162, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.5278, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.8167, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(76.4868, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.0827, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(83.1865, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.3375, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.4892, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.6256, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.1925, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.8768, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.14299242424242425, 'recall': 0.21297602256699577, 'f1': 0.17110481586402268, 'number': 709}, 'P': {'precision': 0.09901873327386262, 'recall': 0.14586070959264127, 'f1': 0.11795961742826781, 'number': 761}, 'overall_precision': 0.12034910427193385, 'overall_recall': 0.1782312925170068, 'overall_f1': 0.1436797367699479, 'overall_accuracy': 0.6750005360550636}
			------------EPOCH 18---------------
Loss:  tensor(79.7780, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(137.9831, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(108.7360, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.3610, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.6642, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.5668, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(165.9188, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.2310, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(113.3476, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(206.1232, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(329.7994, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(89.6679, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(109.1244, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.6439, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.9609, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(69.9192, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.9736, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(241.6674, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(175.3600, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(109.9782, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.8697, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(100.4723, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.2745, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.8424, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.3473, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(75.5394, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.9746, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.0077, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(62.9914, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.8133, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(65.9322, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.8970, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.9060, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.0628, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.1491, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.6444, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.1763157894736842, 'recall': 0.18899858956276447, 'f1': 0.1824370319945541, 'number': 709}, 'P': {'precision': 0.14597544338335608, 'recall': 0.2812089356110381, 'f1': 0.1921867983834755, 'number': 761}, 'overall_precision': 0.15633423180592992, 'overall_recall': 0.23673469387755103, 'overall_f1': 0.18831168831168832, 'overall_accuracy': 0.6686750863048653}
			------------EPOCH 19---------------
Loss:  tensor(81.2609, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(225.5214, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(126.6613, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.9898, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.1406, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.5157, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(198.4640, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.3795, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(141.4091, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(292.4712, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(408.7525, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(74.6664, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(128.9487, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.2971, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.0072, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.8261, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.2337, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(182.7331, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(222.9399, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(128.5320, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.6603, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(92.5831, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.0986, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.7474, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.1329, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(84.1452, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.8813, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.1813, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.1666, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.6327, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(62.1701, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.4030, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.4521, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.8007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.6144, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.3879, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.16499614494988435, 'recall': 0.3018335684062059, 'f1': 0.21335992023928216, 'number': 709}, 'P': {'precision': 0.15177610333692143, 'recall': 0.18528252299605782, 'f1': 0.16686390532544382, 'number': 761}, 'overall_precision': 0.15947888589398024, 'overall_recall': 0.24149659863945577, 'overall_f1': 0.1920995670995671, 'overall_accuracy': 0.6599266676673028}
			------------EPOCH 20---------------
Loss:  tensor(54.1910, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(97.2266, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(101.1211, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.2993, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.9533, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.4144, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(145.9286, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.9227, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(86.3406, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(151.8783, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(289.5411, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(75.3767, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(123.7101, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.2037, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.8380, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.2805, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.4872, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(185.1105, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(120.0454, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(87.3719, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.3185, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(137.1826, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.7558, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.9864, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.5266, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(75.0185, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.5763, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.4816, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.1475, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.4061, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.2118, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.4039, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.1415, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.9820, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.1629, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.6882, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.15986769570011025, 'recall': 0.20451339915373765, 'f1': 0.17945544554455445, 'number': 709}, 'P': {'precision': 0.14992389649923896, 'recall': 0.2588699080157687, 'f1': 0.18987951807228914, 'number': 761}, 'overall_precision': 0.1539846915803692, 'overall_recall': 0.23265306122448978, 'overall_f1': 0.1853156326198862, 'overall_accuracy': 0.6736925616999379}
			------------EPOCH 21---------------
Loss:  tensor(36.3496, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(85.4509, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(74.6267, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.5866, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.4418, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.1924, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(110.7094, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.3885, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(66.0870, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(138.3385, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(205.1295, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.6884, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(66.6372, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.2702, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.5611, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.1489, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.4069, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(146.3691, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(109.0560, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(70.0878, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.8161, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(93.6429, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.0796, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.3078, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.5790, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.9882, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.0649, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.3255, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.1714, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.4288, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.7636, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.3081, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.0275, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.7863, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.0750, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.4656, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.15465898174831894, 'recall': 0.22708039492242596, 'f1': 0.18400000000000002, 'number': 709}, 'P': {'precision': 0.12043189368770764, 'recall': 0.19053876478318002, 'f1': 0.1475826972010178, 'number': 761}, 'overall_precision': 0.13630289532293988, 'overall_recall': 0.20816326530612245, 'overall_f1': 0.16473755047106325, 'overall_accuracy': 0.6738855415228252}
			------------EPOCH 22---------------
Loss:  tensor(31.8410, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(76.4470, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.4492, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.1500, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.2953, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.6358, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(89.9681, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.4655, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.3477, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(118.7382, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(187.1555, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(82.3812, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.6216, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.7387, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.2896, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.2404, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.1064, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(115.6988, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(64.4188, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.9932, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.9871, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(59.6221, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.1627, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.0351, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.4994, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.5280, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.6264, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.8671, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.8548, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.0020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.4907, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.0879, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.1939, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.5022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.5770, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.5688, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.1783893985728848, 'recall': 0.2468265162200282, 'f1': 0.20710059171597633, 'number': 709}, 'P': {'precision': 0.1544058205335489, 'recall': 0.25098554533508544, 'f1': 0.19119119119119118, 'number': 761}, 'overall_precision': 0.16501352569882777, 'overall_recall': 0.24897959183673468, 'overall_f1': 0.19848156182212578, 'overall_accuracy': 0.6792032077535004}
			------------EPOCH 23---------------
Loss:  tensor(26.0878, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(62.4800, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.0134, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.0163, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.1820, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.3162, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(70.3278, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.2048, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.8952, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(83.5014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(110.7406, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.7080, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.7364, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.3661, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.4728, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.5734, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.6895, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(107.3291, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.5089, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.6786, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.2145, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.0985, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.0849, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.0456, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.4035, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.6111, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.2286, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.2073, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.8034, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.8869, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.9308, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.5429, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.0537, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.1594, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.4389, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.2969, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.15367965367965367, 'recall': 0.2002820874471086, 'f1': 0.17391304347826084, 'number': 709}, 'P': {'precision': 0.12965964343598055, 'recall': 0.2102496714848883, 'f1': 0.16040100250626566, 'number': 761}, 'overall_precision': 0.13994439295644115, 'overall_recall': 0.2054421768707483, 'overall_f1': 0.16648291069459759, 'overall_accuracy': 0.6785813838797521}
			------------EPOCH 24---------------
Loss:  tensor(19.8869, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.2571, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.1098, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.9871, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.4303, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.0304, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.8568, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.3807, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.2569, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(98.3847, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(104.7953, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.8703, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.0414, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.0346, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.0742, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.3669, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.3006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(99.1854, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.9923, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.3235, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.3672, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.5741, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.5959, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.8665, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.8492, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.5802, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.7768, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.6509, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.9727, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.3079, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.2960, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.0289, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.1709, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.5612, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.4787, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.8810, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.18858307849133538, 'recall': 0.2609308885754584, 'f1': 0.21893491124260356, 'number': 709}, 'P': {'precision': 0.16613672496025436, 'recall': 0.27463863337713534, 'f1': 0.20703318474492324, 'number': 761}, 'overall_precision': 0.17597141581062975, 'overall_recall': 0.2680272108843537, 'overall_f1': 0.21245618765165813, 'overall_accuracy': 0.6763513948152754}
			------------EPOCH 25---------------
Loss:  tensor(14.8853, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.4909, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.9983, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.3337, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.9389, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.0243, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.6352, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.7232, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.5854, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(68.3382, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(111.8122, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.7021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.5397, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.5912, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.7199, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.2186, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.8979, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(74.4500, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.1931, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.9813, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.9034, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.1423, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.1924, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.2225, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.3234, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.2400, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.7188, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.3860, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.0500, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.7852, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.3537, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.0057, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.3090, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.4451, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.5122, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.0645, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.1579476861167002, 'recall': 0.22143864598025387, 'f1': 0.18438050499119202, 'number': 709}, 'P': {'precision': 0.1285240464344942, 'recall': 0.20367936925098554, 'f1': 0.15760040671072698, 'number': 761}, 'overall_precision': 0.14181818181818182, 'overall_recall': 0.21224489795918366, 'overall_f1': 0.17002724795640325, 'overall_accuracy': 0.6759868773720437}
			------------EPOCH 26---------------
Loss:  tensor(9.4086, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.3438, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.0022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.1006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.1198, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.8918, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.7184, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.4024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.3856, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.5913, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(66.9293, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.8927, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.6026, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.4928, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.4672, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.4915, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.5961, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(66.9815, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.8080, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.5101, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.3891, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.6024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.2519, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.7161, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.1193, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.8343, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.1193, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.4757, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.1822, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.3246, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.6918, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.4471, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.1779, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.9479, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.5926, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.6645, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.18955732122587968, 'recall': 0.23554301833568406, 'f1': 0.210062893081761, 'number': 709}, 'P': {'precision': 0.17114351496546432, 'recall': 0.29303547963206306, 'f1': 0.21608527131782945, 'number': 761}, 'overall_precision': 0.17857142857142858, 'overall_recall': 0.2653061224489796, 'overall_f1': 0.21346469622331693, 'overall_accuracy': 0.6734566974719643}
			------------EPOCH 27---------------
Loss:  tensor(7.1039, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.5325, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.8960, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.8081, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.5261, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.0688, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.6903, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2065, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.4831, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.5516, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.0052, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.5663, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.7247, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.6711, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.6167, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.7572, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.9160, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.8472, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.5807, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.7524, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.6513, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.7068, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5030, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.3469, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.6012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.9640, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.7151, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2934, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.5699, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.7146, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.7210, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.9677, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.0006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.0675, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.5665, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.8848, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.1482213438735178, 'recall': 0.21156558533145275, 'f1': 0.17431725740848344, 'number': 709}, 'P': {'precision': 0.12439024390243902, 'recall': 0.20105124835742444, 'f1': 0.15369161225514816, 'number': 761}, 'overall_precision': 0.13514719000892061, 'overall_recall': 0.20612244897959184, 'overall_f1': 0.16325431034482757, 'overall_accuracy': 0.6751077470677788}
			------------EPOCH 28---------------
Loss:  tensor(5.4495, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.2669, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.0262, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.5569, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.5781, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.6513, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.9014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.8781, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.7119, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.8628, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.0188, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.5449, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.1701, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.9270, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.2097, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.2063, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.5538, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.0366, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.5922, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.5610, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2795, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.0009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.1491, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.0664, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.1526, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.6792, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.3524, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.0179, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.1399, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.8232, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.5846, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.1653, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.0517, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5265, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.8557, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.3841, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.17758046614872364, 'recall': 0.22566995768688294, 'f1': 0.19875776397515527, 'number': 709}, 'P': {'precision': 0.15749235474006115, 'recall': 0.2706964520367937, 'f1': 0.1991300144997583, 'number': 761}, 'overall_precision': 0.1656858306926211, 'overall_recall': 0.24897959183673468, 'overall_f1': 0.19896711062788802, 'overall_accuracy': 0.6724489139524412}
			------------EPOCH 29---------------
Loss:  tensor(4.6158, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.3249, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.0047, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.0690, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.9867, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.4576, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.0737, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5607, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.8828, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.5968, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.0099, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.1703, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.2530, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5919, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.6596, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.7932, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.9946, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.5391, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.0131, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.1740, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.1046, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.7217, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.9017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.8048, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.7481, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.4025, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.0695, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.7412, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.4846, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.3483, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.2446, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.8638, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.8135, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2159, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5097, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.1414, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.16110019646365423, 'recall': 0.231311706629055, 'f1': 0.1899247249565721, 'number': 709}, 'P': {'precision': 0.13666121112929625, 'recall': 0.21944809461235218, 'f1': 0.16843166918809885, 'number': 761}, 'overall_precision': 0.14776785714285715, 'overall_recall': 0.22517006802721087, 'overall_f1': 0.1784366576819407, 'overall_accuracy': 0.6770161030941099}
			------------EPOCH 30---------------
Loss:  tensor(3.8758, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.8948, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.4450, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5852, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.3238, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.7630, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.2026, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3410, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.6373, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.5343, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.6304, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.3408, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.4891, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.1649, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.1211, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.1108, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.7606, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.9830, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.5636, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.2610, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.8131, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.3053, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6509, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6252, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.3773, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.1953, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.7954, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6153, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.5849, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.9795, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.5283, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5234, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.4999, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.9501, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2080, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.9024, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.16310461192350956, 'recall': 0.20451339915373765, 'f1': 0.18147684605757194, 'number': 709}, 'P': {'precision': 0.1419800460475825, 'recall': 0.24310118265440211, 'f1': 0.17926356589147288, 'number': 761}, 'overall_precision': 0.15054744525547445, 'overall_recall': 0.22448979591836735, 'overall_f1': 0.18022938285090112, 'overall_accuracy': 0.6750005360550636}
