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(1383.7799, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1847.1038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2147.7820, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1786.7515, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2207.9121, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1968.4508, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1492.2668, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1376.3262, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1483.4205, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4352.5400, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1697.0276, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1865.6008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2686.7666, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1407.6672, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2581.4431, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1765.3837, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2774.4094, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1875.0990, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1397.2197, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2184.8701, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2457.9871, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2527.3662, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2434.6816, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2056.3479, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3126.1519, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1834.0994, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2219.6431, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2232.8037, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1287.9974, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2393.4219, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1324.6028, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1682.3799, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1518.3234, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1722.8640, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1543.2612, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1984.9557, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1050.7512, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1881.3259, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1573.6863, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2104.2744, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1307.9421, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2145.7798, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1439.9404, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(982.9487, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1894.7828, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1218.4424, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1736.8201, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1143.0654, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1969.8624, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1623.6810, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1539.6759, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1764.6251, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2130.1189, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(743.1539, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(998.1386, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(677.5798, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1697.5679, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(876.1559, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1156.8752, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1101.2070, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.1638418079096045, 'recall': 0.3431952662721893, 'f1': 0.2217973231357553, 'number': 169}, 'P': {'precision': 0.3888888888888889, 'recall': 0.06862745098039216, 'f1': 0.11666666666666667, 'number': 204}, 'overall_precision': 0.18461538461538463, 'overall_recall': 0.19302949061662197, 'overall_f1': 0.18872870249017037, 'overall_accuracy': 0.44048460373548715}
			------------EPOCH 2---------------
Loss:  tensor(905.2062, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1109.9971, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1418.3418, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1180.1002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1577.1633, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1342.6161, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1065.8843, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(970.6295, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1090.9248, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3240.2053, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1244.5011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1305.9644, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1912.0084, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(992.2628, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1821.3347, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1270.5847, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2260.2678, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1417.0005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1036.2122, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1545.2114, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1742.6724, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1929.1188, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2097.2231, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1583.8982, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2350.2349, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1216.1974, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1659.1708, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1867.4474, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1015.2482, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1709.1511, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(893.2701, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1185.4768, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1084.2567, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1376.0759, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1176.0138, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1646.2996, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(737.2122, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1364.0197, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1305.9143, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1655.7524, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1036.3188, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1777.4517, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(905.1281, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(658.9977, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1462.8469, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(908.6422, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1199.4269, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(815.7792, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1405.3729, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1214.0706, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1172.7432, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1136.0676, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1617.4675, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(587.8258, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(812.0640, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(557.7700, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1412.9385, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(753.9856, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1019.1823, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(940.5899, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.23786407766990292, 'recall': 0.28994082840236685, 'f1': 0.2613333333333333, 'number': 169}, 'P': {'precision': 0.3755656108597285, 'recall': 0.4068627450980392, 'f1': 0.3905882352941176, 'number': 204}, 'overall_precision': 0.3091334894613583, 'overall_recall': 0.353887399463807, 'overall_f1': 0.33, 'overall_accuracy': 0.5882887430590611}
			------------EPOCH 3---------------
Loss:  tensor(700.6948, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(885.8530, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1152.3993, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(914.2850, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1297.6313, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1027.1053, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(807.4742, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(752.9077, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(811.4190, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2772.7778, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1018.2058, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1046.1799, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1628.4905, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(780.5732, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1414.9873, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(848.6208, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1995.9070, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1043.0154, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(859.7575, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1275.6772, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1311.3593, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1623.3101, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1456.6917, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1362.6367, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1816.7770, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(919.1145, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1330.0421, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1472.0793, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(944.5122, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1485.6389, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(753.8592, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1077.4537, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(905.8939, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1077.4691, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(941.3068, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1207.3652, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(537.0137, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1010.1693, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1014.2571, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1320.9358, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(858.5788, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1327.0439, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(619.9362, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(452.2918, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1128.5343, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(632.0143, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(637.9263, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(467.9994, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(796.6100, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(980.9653, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(982.7546, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(770.3162, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1104.9940, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(461.6709, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(575.3982, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(439.8062, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1071.1364, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(564.1656, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(776.7432, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(758.4871, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.2946859903381642, 'recall': 0.3609467455621302, 'f1': 0.324468085106383, 'number': 169}, 'P': {'precision': 0.4626865671641791, 'recall': 0.30392156862745096, 'f1': 0.3668639053254438, 'number': 204}, 'overall_precision': 0.36070381231671556, 'overall_recall': 0.3297587131367292, 'overall_f1': 0.3445378151260504, 'overall_accuracy': 0.6485613326602726}
			------------EPOCH 4---------------
Loss:  tensor(505.8060, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(907.1444, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1185.5126, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(752.3556, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(931.0451, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(782.1487, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(518.1130, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(559.2207, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(581.2579, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2322.2036, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(646.3278, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(764.4518, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1066.5979, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(634.5896, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1164.5403, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(490.1734, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1301.0325, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(855.6604, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(706.1283, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(977.7333, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1084.2643, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1655.1089, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1015.3195, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1163.7830, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1323.2117, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(761.2557, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(950.2803, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1000.1544, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(731.3281, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(991.7966, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(499.8984, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(734.4017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(690.4752, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(871.7990, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(688.1554, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(817.8275, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(519.6421, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(920.8920, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(948.0826, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1058.9928, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(945.4049, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1147.3870, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(682.4469, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(487.0832, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(865.0457, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(410.2365, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(430.1417, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(414.2544, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(472.4634, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(798.8160, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(644.9598, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(574.2407, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(716.0709, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(327.0117, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(374.7760, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(329.9848, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(901.2322, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(452.0837, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(659.5867, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(543.0991, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.21122994652406418, 'recall': 0.46745562130177515, 'f1': 0.29097605893186, 'number': 169}, 'P': {'precision': 0.4375, 'recall': 0.1715686274509804, 'f1': 0.2464788732394366, 'number': 204}, 'overall_precision': 0.2511013215859031, 'overall_recall': 0.30563002680965146, 'overall_f1': 0.2756952841596131, 'overall_accuracy': 0.5225643614336194}
			------------EPOCH 5---------------
Loss:  tensor(574.2124, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(663.1539, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(831.0257, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(517.8486, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(905.3589, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(605.8330, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(497.7437, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(523.7261, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(525.6161, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1481.2495, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(432.7150, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(516.4968, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(773.9839, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(530.3655, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(905.8566, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(352.9695, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(925.4537, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1021.3373, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(739.6670, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1149.4047, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(922.8124, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1245.3534, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(608.8545, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1137.4647, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1001.0320, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(633.9246, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(718.4060, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(736.7612, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(681.1445, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(956.8975, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(495.4585, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(470.6886, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(463.6660, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(960.4385, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(599.3244, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(689.4336, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(225.2004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(587.3942, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(522.1555, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(790.4768, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(615.9094, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(976.5753, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(257.7971, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(181.7404, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(531.4426, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(363.1443, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(295.4199, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(332.5396, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(500.3534, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(848.3984, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(630.1017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(406.3322, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(440.1359, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(176.9383, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(226.3380, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(188.4794, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(700.0156, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(306.0434, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(410.5649, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(483.6679, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3745928338762215, 'recall': 0.6804733727810651, 'f1': 0.4831932773109244, 'number': 169}, 'P': {'precision': 0.5087719298245614, 'recall': 0.28431372549019607, 'f1': 0.36477987421383645, 'number': 204}, 'overall_precision': 0.41092636579572445, 'overall_recall': 0.46380697050938335, 'overall_f1': 0.4357682619647355, 'overall_accuracy': 0.6571428571428571}
			------------EPOCH 6---------------
Loss:  tensor(214.9240, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(361.4300, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(438.0300, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(275.0277, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(640.9747, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(498.4237, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(367.6091, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(445.8622, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(460.6115, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1414.6931, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(444.2955, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(464.2440, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(848.9133, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(354.1998, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(614.3046, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(347.1307, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(647.3940, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(489.7916, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(332.8845, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(594.2908, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(430.7709, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(604.3033, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(363.1454, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(610.8336, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1104.4291, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(713.4131, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(984.3651, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(708.7062, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(833.7144, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(736.4303, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(419.1788, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(634.9695, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(566.6404, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(540.6194, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(438.9981, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(624.2587, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(143.4381, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(427.3064, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(362.1354, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(502.9760, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(438.8080, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(543.2549, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(269.6642, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(171.8228, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(675.2762, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(273.7947, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(244.1302, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(157.5656, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(242.1783, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(860.5868, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(816.1224, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(525.1466, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(449.2523, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(212.8528, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(219.8384, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(246.1805, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(587.5455, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(361.9109, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(420.0710, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(288.5533, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.39649122807017545, 'recall': 0.6686390532544378, 'f1': 0.49779735682819387, 'number': 169}, 'P': {'precision': 0.56875, 'recall': 0.44607843137254904, 'f1': 0.5, 'number': 204}, 'overall_precision': 0.4584269662921348, 'overall_recall': 0.546916890080429, 'overall_f1': 0.49877750611246935, 'overall_accuracy': 0.6764260474507824}
			------------EPOCH 7---------------
Loss:  tensor(166.4989, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(257.1830, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(310.0281, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(222.3030, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(390.9296, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(294.8283, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(173.6642, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(230.0041, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(254.3839, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(846.1214, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(195.3926, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(217.0053, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(457.7978, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(254.6909, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(447.8394, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(154.0961, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(472.5452, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(326.3202, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(228.9146, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(318.2479, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(322.9149, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(382.9451, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(225.0119, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(541.8519, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(352.7632, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(268.1003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(361.0121, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(236.0128, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(365.5626, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(398.9145, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(177.6311, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(189.5062, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(265.4820, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(269.5971, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(281.4524, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(311.6244, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(164.4303, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(384.3076, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(283.0020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(438.3546, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(433.2726, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(467.6792, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(153.1496, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(127.0499, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(261.9695, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(164.8634, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(168.1071, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(131.4157, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(116.3993, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(303.4593, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(251.5294, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(123.6777, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(152.6460, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(84.4358, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(80.7051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(115.0462, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(268.1387, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(183.3629, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(266.8965, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(233.2000, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3987730061349693, 'recall': 0.7692307692307693, 'f1': 0.5252525252525252, 'number': 169}, 'P': {'precision': 0.640625, 'recall': 0.20098039215686275, 'f1': 0.30597014925373134, 'number': 204}, 'overall_precision': 0.43846153846153846, 'overall_recall': 0.4584450402144772, 'overall_f1': 0.4482306684141546, 'overall_accuracy': 0.6586572438162545}
			------------EPOCH 8---------------
Loss:  tensor(105.3810, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(257.2720, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(222.0459, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(181.8091, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(371.8729, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(337.8803, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(159.9253, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(190.2792, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(186.7661, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(493.5872, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(159.2285, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(168.4496, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(248.5229, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(254.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(303.5132, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(106.6875, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(257.5738, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(216.6722, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(160.5514, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(168.8595, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(119.1354, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(180.0157, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(67.6405, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(255.6501, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(196.8391, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(154.4673, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(228.3504, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(90.3729, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(206.2643, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(326.8926, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(96.2687, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(102.3144, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(160.8988, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(219.5515, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(229.7722, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(218.7095, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(62.9950, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(230.7119, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(209.0732, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(374.5313, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(214.1108, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(408.3377, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.6653, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(79.1341, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(265.4944, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(190.2356, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(110.1162, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(66.7934, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(113.7293, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(220.7564, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(168.5445, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(107.6442, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(112.5244, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.9778, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(64.6376, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(72.6731, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(168.2463, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(149.4134, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(232.6631, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(180.9835, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4280442804428044, 'recall': 0.6863905325443787, 'f1': 0.5272727272727272, 'number': 169}, 'P': {'precision': 0.6176470588235294, 'recall': 0.4117647058823529, 'f1': 0.4941176470588236, 'number': 204}, 'overall_precision': 0.4914004914004914, 'overall_recall': 0.5361930294906166, 'overall_f1': 0.5128205128205129, 'overall_accuracy': 0.6897526501766784}
			------------EPOCH 9---------------
Loss:  tensor(81.5488, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(126.1649, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(113.6038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(129.5027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(171.5387, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(154.4918, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(83.9114, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(92.5216, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(129.4720, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(438.3024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(104.3877, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(107.1171, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(335.8184, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(158.1967, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(376.2796, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(126.0064, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(220.2270, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(330.2021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(212.2842, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(200.0062, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(129.0257, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(129.5929, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(103.2055, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(245.3096, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(106.1364, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(173.9992, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(176.5208, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(97.8519, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(194.9439, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(309.5553, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(89.1652, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(103.8478, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(174.8670, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(174.8803, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(237.6519, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(178.2423, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(104.4959, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(147.3910, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(147.3136, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(151.8364, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(163.8281, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(289.6693, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(77.7485, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(61.5303, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(173.3104, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(79.0897, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(123.6590, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(76.9978, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(131.8035, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(173.4050, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(265.5086, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(154.9827, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(314.1258, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(187.3117, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(132.7769, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(79.4435, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(250.6445, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(145.9484, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(205.4794, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(183.4626, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.37272727272727274, 'recall': 0.727810650887574, 'f1': 0.49298597194388777, 'number': 169}, 'P': {'precision': 0.5723270440251572, 'recall': 0.44607843137254904, 'f1': 0.5013774104683196, 'number': 204}, 'overall_precision': 0.4376278118609407, 'overall_recall': 0.5737265415549598, 'overall_f1': 0.4965197215777263, 'overall_accuracy': 0.629984856133266}
			------------EPOCH 10---------------
Loss:  tensor(143.4997, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(138.9947, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(210.6388, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.4111, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(141.9857, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(98.2169, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(81.9895, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(84.0126, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(122.9873, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(264.9660, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(78.8907, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(67.6189, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(143.0072, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(110.9489, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(174.0019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.0769, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(154.6358, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(162.3762, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(118.9327, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(120.5854, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(201.9919, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(272.2949, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(67.0824, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(180.9442, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(164.2831, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(184.3102, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(358.3232, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(190.4347, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(209.7215, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(343.5471, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(97.8482, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(109.0478, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(155.5086, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(208.0919, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(198.8863, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(386.1777, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(62.0115, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(113.8400, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(194.0287, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(176.7785, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(175.5439, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(142.0835, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.7397, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.1395, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(155.9602, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(64.1416, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(94.7890, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(111.8075, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(296.8691, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(425.0102, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(342.2853, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(206.5710, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(223.3665, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(89.2731, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(162.3316, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(109.8060, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(237.6880, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(147.0344, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(164.2867, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(234.1848, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.24681933842239187, 'recall': 0.5739644970414202, 'f1': 0.34519572953736655, 'number': 169}, 'P': {'precision': 0.6373626373626373, 'recall': 0.28431372549019607, 'f1': 0.3932203389830508, 'number': 204}, 'overall_precision': 0.3202479338842975, 'overall_recall': 0.4155495978552279, 'overall_f1': 0.3617269544924154, 'overall_accuracy': 0.5332660272589601}
			------------EPOCH 11---------------
Loss:  tensor(222.2703, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(353.0614, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(485.7382, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(193.6078, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(865.5837, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(353.9966, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(405.4464, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(265.5255, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(467.4500, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(853.7806, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(268.0467, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(251.0806, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(329.7788, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(244.6593, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(325.1974, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(192.1874, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(183.2060, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(93.7787, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(64.1941, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(99.4178, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.2556, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(85.8643, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.0179, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(93.9371, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(108.8875, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(143.4594, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(150.7493, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(74.3073, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(158.5439, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(280.2504, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(103.2095, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(146.9921, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(250.9989, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(265.3497, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(358.9598, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(235.8783, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(225.2753, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(306.6619, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(346.5301, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(321.2117, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(351.1829, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(487.0464, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(130.8276, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(159.1427, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(191.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.4849, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(126.9547, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(83.6482, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(61.2886, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(182.8088, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(158.7792, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(85.9291, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(96.9558, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(74.4502, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(83.1415, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(59.5567, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(225.1593, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(158.9319, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(92.4995, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(162.0849, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4336283185840708, 'recall': 0.5798816568047337, 'f1': 0.49620253164556966, 'number': 169}, 'P': {'precision': 0.48148148148148145, 'recall': 0.5735294117647058, 'f1': 0.5234899328859061, 'number': 204}, 'overall_precision': 0.4584221748400853, 'overall_recall': 0.5764075067024129, 'overall_f1': 0.5106888361045131, 'overall_accuracy': 0.6431095406360424}
			------------EPOCH 12---------------
Loss:  tensor(48.2724, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(160.8411, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(306.9109, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(274.0598, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(261.0195, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(101.7359, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.4888, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.3783, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(97.3629, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(295.7860, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(73.6857, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(84.7540, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(235.8018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(139.5882, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(317.9723, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(164.6718, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(624.8267, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(341.0151, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(213.1033, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(380.3092, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(169.2095, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(371.6864, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(289.9731, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(401.0261, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(104.6284, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(139.7169, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(192.5156, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(124.0211, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(137.5663, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(291.8164, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(97.5451, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(101.6648, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(68.9190, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(92.4086, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(138.0496, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(89.3541, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.5447, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(79.2256, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(72.6227, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.1810, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(115.7600, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(132.8390, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.3701, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.1831, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(77.8631, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(81.1257, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(80.5382, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(77.5139, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(95.6557, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(235.3932, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(148.8298, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(101.6180, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(271.5964, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(102.7643, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(136.1594, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.5212, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(576.1427, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(389.3113, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(398.2627, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(812.0459, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5031847133757962, 'recall': 0.46745562130177515, 'f1': 0.48466257668711654, 'number': 169}, 'P': {'precision': 0.5238095238095238, 'recall': 0.6470588235294118, 'f1': 0.5789473684210527, 'number': 204}, 'overall_precision': 0.5158924205378973, 'overall_recall': 0.5656836461126006, 'overall_f1': 0.5396419437340154, 'overall_accuracy': 0.6718828874305907}
			------------EPOCH 13---------------
Loss:  tensor(103.6929, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(97.6532, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(244.3501, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(148.6443, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(280.1284, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(96.6148, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.0991, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.9299, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(83.6904, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(252.8610, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(70.7108, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.0538, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(69.7407, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(127.0911, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(147.5944, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(157.6625, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(90.3376, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(192.9429, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(95.2386, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(75.6312, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(268.4970, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(232.2249, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(315.4161, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(537.0505, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(424.2261, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(284.8963, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(397.0865, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(319.7596, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(139.2414, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(259.3967, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(73.8784, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.2079, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(62.4551, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(95.7041, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(124.5674, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(123.6873, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.5832, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(75.3852, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(136.3324, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(159.5996, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(155.0294, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(183.0524, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.9000, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.3488, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(114.0624, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(105.8832, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(90.9478, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(77.1257, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(298.9878, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(173.9331, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(122.3988, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(110.0137, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(153.7363, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.3164, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(213.1628, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(102.6135, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(386.1196, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(149.7724, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(424.6024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(412.2590, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5217391304347826, 'recall': 0.35502958579881655, 'f1': 0.4225352112676056, 'number': 169}, 'P': {'precision': 0.4523809523809524, 'recall': 0.7450980392156863, 'f1': 0.562962962962963, 'number': 204}, 'overall_precision': 0.4700665188470067, 'overall_recall': 0.5683646112600537, 'overall_f1': 0.5145631067961165, 'overall_accuracy': 0.62221100454316}
			------------EPOCH 14---------------
Loss:  tensor(64.1496, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(146.0614, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(159.0766, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(183.1847, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(215.9996, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(235.1868, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(381.9645, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(268.6619, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(110.9598, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(255.9920, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(85.5087, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(178.8326, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.2767, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(79.4723, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(88.0167, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.7250, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.7928, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(87.2288, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(77.6612, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.3986, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(74.0603, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(73.7189, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.5352, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(113.5349, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(364.4646, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(413.6448, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(862.8990, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(376.0368, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(433.9981, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1026.1677, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(891.2262, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(450.6862, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(156.2197, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(602.6556, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(530.8405, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(263.7809, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(80.5498, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(199.2467, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(300.2241, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(508.6745, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(89.1834, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(80.1928, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(168.4941, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(82.8666, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(68.1682, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.1970, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.2264, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.3437, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.8394, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(89.1791, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(99.8534, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.0716, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(122.3980, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.7953, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(93.5136, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.7836, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(513.9222, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(121.0739, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(344.8116, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(317.8022, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3, 'recall': 0.08875739644970414, 'f1': 0.136986301369863, 'number': 169}, 'P': {'precision': 0.3826879271070615, 'recall': 0.8235294117647058, 'f1': 0.5225505443234837, 'number': 204}, 'overall_precision': 0.37423312883435583, 'overall_recall': 0.4906166219839142, 'overall_f1': 0.42459396751740147, 'overall_accuracy': 0.5764765270065624}
			------------EPOCH 15---------------
Loss:  tensor(443.5778, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1066.6123, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(983.8816, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1039.1230, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1467.4626, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(716.7134, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(813.5625, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1102.6086, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(554.4255, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2729.2446, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(412.7236, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(531.2962, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(345.8694, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(190.7155, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(524.4718, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(85.6186, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(79.0780, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(124.6715, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(67.3161, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(70.0919, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(92.8622, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(73.0602, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.3380, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(113.9326, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(719.2884, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(274.1848, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(340.7411, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(186.3319, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(291.5219, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1091.6727, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(425.8080, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(283.7535, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(508.9501, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2006.6335, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(479.6601, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1137.6753, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(512.5874, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1302.2850, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1468.7024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1527.6538, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(389.1925, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(556.1888, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(470.1318, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(142.3763, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(587.2503, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(169.4944, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(157.2268, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(109.5167, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(77.8851, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(305.5196, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(278.3005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(246.2138, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(169.8527, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(119.6059, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(101.4343, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(99.7506, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(231.4151, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(127.3160, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(156.3769, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(178.3201, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4083333333333333, 'recall': 0.28994082840236685, 'f1': 0.3391003460207612, 'number': 169}, 'P': {'precision': 0.4676470588235294, 'recall': 0.7794117647058824, 'f1': 0.5845588235294118, 'number': 204}, 'overall_precision': 0.45217391304347826, 'overall_recall': 0.5576407506702413, 'overall_f1': 0.4993997599039616, 'overall_accuracy': 0.6325088339222615}
			------------EPOCH 16---------------
Loss:  tensor(123.6461, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(294.3523, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(235.8739, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(185.7217, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(499.4460, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(317.4941, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(319.5275, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(307.7239, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(429.5995, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1345.3250, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(494.6508, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(575.1964, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(696.6079, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(471.2904, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(867.7057, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(240.2750, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(723.0247, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(934.5770, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(797.0820, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1082.3964, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(844.3108, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(813.7037, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(584.4218, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(999.7161, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(635.7867, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(205.9760, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(306.8125, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(428.7195, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(331.5809, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(348.5599, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(128.9073, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(138.6131, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(180.5021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(169.6005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(177.0342, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(108.4226, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(98.0842, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(306.5373, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(214.4033, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(252.1087, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(191.1634, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(334.7743, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(158.7270, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(88.0282, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(538.4463, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(196.5214, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(192.5519, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(138.9998, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(178.7056, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(699.1273, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(629.4758, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(645.5460, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(622.6489, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(284.6642, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(309.8662, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(289.5400, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(524.2604, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(409.7245, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(402.6204, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(278.6779, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.40236686390532544, 'recall': 0.8047337278106509, 'f1': 0.5364891518737673, 'number': 169}, 'P': {'precision': 0.8051948051948052, 'recall': 0.30392156862745096, 'f1': 0.44128113879003555, 'number': 204}, 'overall_precision': 0.4771084337349398, 'overall_recall': 0.5308310991957105, 'overall_f1': 0.5025380710659899, 'overall_accuracy': 0.6664310954063605}
			------------EPOCH 17---------------
Loss:  tensor(170.1011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(197.9799, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(445.3792, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(214.4666, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(310.2389, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(250.7863, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(125.4164, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(192.9858, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(96.2956, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(498.2044, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(120.1787, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(156.4308, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(159.8515, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(110.4855, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(191.4575, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(88.3243, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(88.6528, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(110.4423, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(83.1955, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(70.3667, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(105.7941, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(71.2524, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.5128, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(240.4505, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(155.2935, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(86.7216, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(115.1666, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(62.5442, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(277.9695, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(181.8005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(74.0816, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(71.5354, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(137.9059, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(97.0820, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(141.1606, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(82.2059, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.4875, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(240.5444, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(124.9648, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(153.3027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(145.0269, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(151.8063, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.8758, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.6592, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(135.3958, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.6774, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.4051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.3320, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.3933, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(74.9924, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(82.2698, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.4155, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.9145, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.7969, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.8067, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.8178, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.8938, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.9189, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(81.0251, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.8595, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.518324607329843, 'recall': 0.5857988165680473, 'f1': 0.55, 'number': 169}, 'P': {'precision': 0.5778688524590164, 'recall': 0.6911764705882353, 'f1': 0.6294642857142857, 'number': 204}, 'overall_precision': 0.5517241379310345, 'overall_recall': 0.6434316353887399, 'overall_f1': 0.594059405940594, 'overall_accuracy': 0.7150933871781928}
			------------EPOCH 18---------------
Loss:  tensor(14.4985, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.1347, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.0142, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.3197, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.1511, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.6411, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.1975, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.3216, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.7420, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(166.4231, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.8803, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.6480, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.7837, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(59.1759, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.7971, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.3820, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.0960, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.5935, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.0716, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.6458, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.6906, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.8894, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.2251, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(86.1043, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.3853, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(79.5811, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.3746, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.2519, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(66.7817, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(125.6233, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.8531, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.7630, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.8404, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.8900, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.7274, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.5585, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.5473, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.1137, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.7024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.8899, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.7655, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.5233, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.9355, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.8857, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(67.2563, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.9767, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.5037, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.8186, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.5224, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.7395, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.1709, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.8868, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.7228, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.2322, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.9835, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.4995, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.5335, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(66.4756, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.3645, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.2975, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4977578475336323, 'recall': 0.6568047337278107, 'f1': 0.566326530612245, 'number': 169}, 'P': {'precision': 0.6403940886699507, 'recall': 0.6372549019607843, 'f1': 0.6388206388206389, 'number': 204}, 'overall_precision': 0.5657276995305164, 'overall_recall': 0.646112600536193, 'overall_f1': 0.6032540675844806, 'overall_accuracy': 0.7373043917213529}
			------------EPOCH 19---------------
Loss:  tensor(8.6745, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.1242, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.9465, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.6438, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.3021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.9350, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.9977, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.2002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.2090, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(86.3996, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.4305, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.5740, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.6297, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.1772, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.7002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.9767, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.4821, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.1866, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.8479, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.7865, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.9662, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.5191, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.0072, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.0562, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.9765, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.5260, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.2347, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.7440, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.4163, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(93.3688, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.6272, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.9635, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.7373, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.8330, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.8898, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.2035, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.5453, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.3062, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.4435, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.3592, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.1541, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.3890, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.0232, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.8684, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.7974, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.5612, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.5571, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.5003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.3680, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.4361, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.5774, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.4984, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.6348, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.9734, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.8918, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.2671, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.8198, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.0720, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.6976, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.3954, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.49344978165938863, 'recall': 0.6686390532544378, 'f1': 0.5678391959798995, 'number': 169}, 'P': {'precision': 0.6354679802955665, 'recall': 0.6323529411764706, 'f1': 0.6339066339066338, 'number': 204}, 'overall_precision': 0.5601851851851852, 'overall_recall': 0.6487935656836461, 'overall_f1': 0.6012422360248448, 'overall_accuracy': 0.732761231701161}
			------------EPOCH 20---------------
Loss:  tensor(6.5209, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.5724, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.3304, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.2617, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.4789, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.0170, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.5101, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.5009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.1226, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(59.8175, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.2695, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.3862, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.0524, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.3445, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.7735, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.3137, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.7975, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.9380, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.9720, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.4999, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.4348, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.0509, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.5452, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.1347, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.6699, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.9172, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.5745, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.3225, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.2321, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(79.7593, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.7191, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.9118, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.1726, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.5265, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.4063, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.9808, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.7028, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.8177, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.7199, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.3124, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.9880, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.2612, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.8456, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.0559, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.0046, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.4459, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.5986, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.4738, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.6434, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.2737, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.8462, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.1901, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.4146, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.8931, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.1346, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.7340, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.0273, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.0424, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.3928, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.3796, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4890829694323144, 'recall': 0.6627218934911243, 'f1': 0.5628140703517588, 'number': 169}, 'P': {'precision': 0.625, 'recall': 0.6372549019607843, 'f1': 0.6310679611650485, 'number': 204}, 'overall_precision': 0.5537757437070938, 'overall_recall': 0.6487935656836461, 'overall_f1': 0.5975308641975309, 'overall_accuracy': 0.7257950530035335}
			------------EPOCH 21---------------
Loss:  tensor(5.6856, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.2036, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.4348, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.7334, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.0154, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.9720, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.1945, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.4724, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.4103, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.8026, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.9572, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.6944, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.7519, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.4854, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.4760, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.4952, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.8751, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.9592, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.2308, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.2897, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.8617, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.0079, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.7238, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.5932, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.3559, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.9544, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.0410, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.0605, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.0698, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(72.7758, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.0513, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.1810, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.4920, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.1371, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.4037, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.1277, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.5508, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.0256, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.5324, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.3598, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.1826, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.2458, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.7049, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.4610, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.3711, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.9050, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.7578, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.6229, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.6262, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.3431, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.1536, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.7994, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.8857, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.8802, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.9296, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.4906, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.7619, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.2043, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.5366, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.1827, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.49122807017543857, 'recall': 0.6627218934911243, 'f1': 0.5642317380352645, 'number': 169}, 'P': {'precision': 0.6220095693779905, 'recall': 0.6372549019607843, 'f1': 0.6295399515738499, 'number': 204}, 'overall_precision': 0.5537757437070938, 'overall_recall': 0.6487935656836461, 'overall_f1': 0.5975308641975309, 'overall_accuracy': 0.7252902574457345}
			------------EPOCH 22---------------
Loss:  tensor(4.9628, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.3970, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.6410, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.5702, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.1841, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.4433, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.7726, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.6131, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.5102, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.3640, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.0728, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.3197, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.0355, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.3152, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.4718, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.6162, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.4717, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.7757, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.1374, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.4067, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.4100, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.7683, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.1781, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.5742, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.2563, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.0401, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.9761, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.1907, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.4006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(67.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.0076, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.9570, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.1769, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.0993, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.5551, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.6505, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.7798, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.5389, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.0057, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.4210, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.8227, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.4814, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.9406, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.9668, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.0652, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5436, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.3496, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.2586, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.0610, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.0158, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.6910, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.9915, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.6454, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.2390, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.2889, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.4593, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.0900, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.6706, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.4064, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.7825, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4977973568281938, 'recall': 0.6686390532544378, 'f1': 0.5707070707070707, 'number': 169}, 'P': {'precision': 0.6220095693779905, 'recall': 0.6372549019607843, 'f1': 0.6295399515738499, 'number': 204}, 'overall_precision': 0.5573394495412844, 'overall_recall': 0.6514745308310992, 'overall_f1': 0.6007416563658838, 'overall_accuracy': 0.7253912165572943}
			------------EPOCH 23---------------
Loss:  tensor(4.4244, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.7271, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.9472, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.5811, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.9978, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.1456, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.7771, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.2210, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.8666, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.8588, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.4555, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.3621, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.5844, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.5362, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.2730, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.3609, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.2201, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.7373, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.0400, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.3989, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.6887, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.9548, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.7813, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.1746, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.5390, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.4995, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.6073, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.5383, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.7956, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(61.7771, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.5209, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.1633, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.7530, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.6825, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.3297, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.6715, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.2926, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.6250, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.1972, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.0438, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.0690, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.2884, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.3379, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.6439, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.1215, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2864, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.2203, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.2330, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.6398, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.2091, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.6639, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.4733, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.1296, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.8850, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.0606, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.7359, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.6374, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.7284, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.5787, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.6911, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5022222222222222, 'recall': 0.6686390532544378, 'f1': 0.5736040609137056, 'number': 169}, 'P': {'precision': 0.6113744075829384, 'recall': 0.6323529411764706, 'f1': 0.6216867469879518, 'number': 204}, 'overall_precision': 0.555045871559633, 'overall_recall': 0.6487935656836461, 'overall_f1': 0.5982694684796044, 'overall_accuracy': 0.7235739525492175}
			------------EPOCH 24---------------
Loss:  tensor(3.8277, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.5780, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.9515, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.9839, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.1513, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.9445, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.9296, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.1591, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.3033, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.5542, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.8711, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.6486, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.8606, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.2317, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.0371, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.8370, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.9526, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.3608, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.0677, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.3474, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.1113, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.2752, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.4598, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.1587, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.0464, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.3026, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.4953, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.0063, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.4792, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.4997, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.2735, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.8687, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.3298, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.3630, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.2385, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.9199, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.7833, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.8130, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.6928, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.6601, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.2581, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.2720, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.7860, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.3410, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.1960, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.0387, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.1536, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.4601, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.0966, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.5485, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.7019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.9663, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.0456, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.9096, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.2558, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.1852, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.4234, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.2681, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.9019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.6518, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5022222222222222, 'recall': 0.6686390532544378, 'f1': 0.5736040609137056, 'number': 169}, 'P': {'precision': 0.6084905660377359, 'recall': 0.6323529411764706, 'f1': 0.6201923076923076, 'number': 204}, 'overall_precision': 0.5537757437070938, 'overall_recall': 0.6487935656836461, 'overall_f1': 0.5975308641975309, 'overall_accuracy': 0.723372034326098}
			------------EPOCH 25---------------
Loss:  tensor(3.3619, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.6706, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.0120, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.6492, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.4760, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.7358, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.2622, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.4238, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.6256, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.2254, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.4510, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.0033, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.2476, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.8770, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.9545, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.3421, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.7187, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.0157, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.1253, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.5751, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.6442, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.6825, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.1574, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.3457, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.6397, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.3951, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.6610, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.5260, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.9607, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.7721, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.9214, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.9612, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.2397, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.4801, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.2978, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.8935, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.2422, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.0476, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.2586, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.7514, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.5383, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.4339, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.3647, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.0491, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.5039, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.8184, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.2446, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.8811, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.5966, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.1790, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.8712, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.5208, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.2742, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.4538, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.7050, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.7772, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.4369, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.3500, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.2928, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.9062, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5022222222222222, 'recall': 0.6686390532544378, 'f1': 0.5736040609137056, 'number': 169}, 'P': {'precision': 0.6113744075829384, 'recall': 0.6323529411764706, 'f1': 0.6216867469879518, 'number': 204}, 'overall_precision': 0.555045871559633, 'overall_recall': 0.6487935656836461, 'overall_f1': 0.5982694684796044, 'overall_accuracy': 0.721655729429581}
			------------EPOCH 26---------------
Loss:  tensor(3.1217, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.9749, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.2890, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.2023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.9391, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.8077, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.6996, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.8511, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.7889, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.9620, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.0068, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.4723, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.7281, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.0100, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.0295, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.8071, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.4668, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.9373, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.2504, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.9031, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.2626, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.2248, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.9280, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.6639, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.2138, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.5544, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.0313, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.2165, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.7885, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.6316, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.5793, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.3306, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.4010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.9200, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.0404, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.2224, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.8215, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.5484, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.8830, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.4324, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.8166, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.7696, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.1095, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.8159, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.1159, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6413, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.4940, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.4590, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.2100, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.1397, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.1496, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.1600, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.6351, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.3114, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.2328, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.4398, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.5353, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.7238, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.7824, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.2088, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5044642857142857, 'recall': 0.6686390532544378, 'f1': 0.5750636132315521, 'number': 169}, 'P': {'precision': 0.6056338028169014, 'recall': 0.6323529411764706, 'f1': 0.618705035971223, 'number': 204}, 'overall_precision': 0.5537757437070938, 'overall_recall': 0.6487935656836461, 'overall_f1': 0.5975308641975309, 'overall_accuracy': 0.7218576476527007}
			------------EPOCH 27---------------
Loss:  tensor(2.8219, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.4529, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.7374, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.6397, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.5041, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.0968, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.2972, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.4667, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.2009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.9485, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.7259, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.0827, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.3489, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.4565, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.3633, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5265, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.4730, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.1375, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.6028, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.3175, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.9827, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.8809, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.7314, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.8516, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.8735, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.8216, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.5199, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.8997, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.9803, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.4661, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.8226, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.7931, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.7331, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.3155, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.0203, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.6008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.4660, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.0361, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.5089, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.4338, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.0474, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.2495, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.8456, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6478, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.7722, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5124, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.8836, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.1002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.9520, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.3256, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.5820, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.9168, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.1300, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.2151, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.8275, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.1556, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.6221, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.3619, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.3208, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.5896, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5022026431718062, 'recall': 0.6745562130177515, 'f1': 0.5757575757575757, 'number': 169}, 'P': {'precision': 0.6113744075829384, 'recall': 0.6323529411764706, 'f1': 0.6216867469879518, 'number': 204}, 'overall_precision': 0.5547945205479452, 'overall_recall': 0.6514745308310992, 'overall_f1': 0.5992601726263872, 'overall_accuracy': 0.7186269560827865}
			------------EPOCH 28---------------
Loss:  tensor(2.6450, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.0670, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.3099, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.0412, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.1592, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.4640, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.7684, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.0186, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.2881, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.9116, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.4386, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.7808, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.0363, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.0157, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.7807, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.3254, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.7429, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.6141, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.8545, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.8330, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.7527, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5979, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6035, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.3383, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.7480, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.8109, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.0561, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.6162, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.3421, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.1671, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.5921, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.3717, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.2124, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.8602, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.2811, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.1110, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.1469, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.5435, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.9939, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.7143, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.2825, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.7729, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.6563, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4935, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.2698, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3877, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.3708, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.8109, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.7033, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.6700, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.1803, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.6566, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.6673, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.4684, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.4629, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.8802, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.8598, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.3440, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.8341, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.0670, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5022026431718062, 'recall': 0.6745562130177515, 'f1': 0.5757575757575757, 'number': 169}, 'P': {'precision': 0.6095238095238096, 'recall': 0.6274509803921569, 'f1': 0.6183574879227053, 'number': 204}, 'overall_precision': 0.5537757437070938, 'overall_recall': 0.6487935656836461, 'overall_f1': 0.5975308641975309, 'overall_accuracy': 0.719434628975265}
			------------EPOCH 29---------------
Loss:  tensor(2.3127, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.7117, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.9740, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.1882, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.8618, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.9725, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.4229, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.7242, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.1552, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.8782, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2333, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.4765, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.7558, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.6822, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.3046, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.0848, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.0745, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.1601, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.2516, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.4278, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5226, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.3655, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4485, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.8490, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.0508, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.6975, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.6602, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.3662, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.8419, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.0595, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.5184, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.0283, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.8266, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.4007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.6716, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.7633, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.8320, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.8720, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.3940, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.0756, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.5322, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.3603, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.4344, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3855, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.9871, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2945, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.9508, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5426, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5517, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.1017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.1447, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5181, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.2864, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.5582, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.1424, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.6650, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.9742, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.4416, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.3945, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.4394, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5022026431718062, 'recall': 0.6745562130177515, 'f1': 0.5757575757575757, 'number': 169}, 'P': {'precision': 0.6124401913875598, 'recall': 0.6274509803921569, 'f1': 0.6198547215496367, 'number': 204}, 'overall_precision': 0.555045871559633, 'overall_recall': 0.6487935656836461, 'overall_f1': 0.5982694684796044, 'overall_accuracy': 0.7179202423018678}
			------------EPOCH 30---------------
Loss:  tensor(2.0174, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.4569, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.6890, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.0504, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.6024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.5417, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.9922, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.3703, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.6112, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.4430, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.0285, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2731, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5389, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.4110, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.8978, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.9488, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.4973, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.8568, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.5942, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.1273, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.3514, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.1658, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3574, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.7869, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.5723, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.5436, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.2572, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.1099, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.4191, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.6560, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.0466, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.7431, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5132, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.0623, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.1924, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.5136, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.5066, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.0948, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.1703, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.5499, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.8364, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.0280, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.2771, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2695, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.7548, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2053, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5992, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3271, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.3925, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.6282, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.3992, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.3439, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.9602, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.8870, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.8701, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.4637, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.1552, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.7524, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.9220, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.8901, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5198237885462555, 'recall': 0.6982248520710059, 'f1': 0.595959595959596, 'number': 169}, 'P': {'precision': 0.6095238095238096, 'recall': 0.6274509803921569, 'f1': 0.6183574879227053, 'number': 204}, 'overall_precision': 0.562929061784897, 'overall_recall': 0.6595174262734584, 'overall_f1': 0.6074074074074074, 'overall_accuracy': 0.7188288743059061}


		-------------RUN 2-----------
			------------EPOCH 1---------------
Loss:  tensor(1782.4347, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2903.5200, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3804.0146, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1907.3666, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2770.6245, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4424.0264, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3036.1689, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2513.9688, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1504.8174, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1707.4580, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2289.8389, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2080.8828, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1205.0298, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(930.2272, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1935.2682, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2469.5552, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1400.3273, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2591.4946, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2946.8818, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2261.9683, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1458.3341, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1597.5784, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(615.0797, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1759.1334, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2624.1240, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1500.7737, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(938.2289, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1925.2990, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1592.8782, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1155.4633, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1931.4250, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2894.0967, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2636.2795, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2141.6089, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1063.5687, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1309.7908, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1355.7909, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1381.5637, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2011.9158, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1649.5137, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1412.6101, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1582.5759, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1676.8960, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1734.3145, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1868.6558, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2885.6438, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1674.5767, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1636.6343, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2065.7446, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1183.0861, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(711.4882, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(976.9624, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1693.6187, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2459.8701, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2092.6794, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2310.6250, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1752.3081, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1548.7098, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.15555555555555556, 'recall': 0.05166051660516605, 'f1': 0.0775623268698061, 'number': 271}, 'P': {'precision': 0.20397111913357402, 'recall': 0.3087431693989071, 'f1': 0.2456521739130435, 'number': 366}, 'overall_precision': 0.1972049689440994, 'overall_recall': 0.19937205651491366, 'overall_f1': 0.19828259172521467, 'overall_accuracy': 0.5722721276462089}
			------------EPOCH 2---------------
Loss:  tensor(1107.8687, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1813.8198, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2648.9651, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1214.0281, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2193.4565, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3110.2622, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2055.5623, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1677.1740, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1067.0931, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1141.7893, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1457.2306, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1370.2321, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(954.4993, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(763.4061, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1510.0253, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1977.7262, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1145.9653, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2088.9124, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2419.5027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1819.1057, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1170.2229, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1334.9237, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(521.4773, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1515.0396, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2307.5820, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1268.0431, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(736.0211, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1582.9109, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1247.5098, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(984.4329, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1638.4332, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2492.5017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2246.5591, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1741.8673, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(870.6801, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1162.9167, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1131.0040, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1154.1279, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1686.0111, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1369.7454, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1236.1102, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1518.5869, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1555.8596, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1547.8428, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1414.8459, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2646.7085, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1323.4028, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1415.9160, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1611.1619, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(994.1354, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(581.4509, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(832.2714, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1335.5267, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1845.9852, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1611.5172, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1696.0581, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1331.1831, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1226.4343, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.08839779005524862, 'recall': 0.05904059040590406, 'f1': 0.07079646017699115, 'number': 271}, 'P': {'precision': 0.20093457943925233, 'recall': 0.3524590163934426, 'f1': 0.25595238095238093, 'number': 366}, 'overall_precision': 0.17618469015795868, 'overall_recall': 0.22762951334379905, 'overall_f1': 0.19863013698630136, 'overall_accuracy': 0.5971480620642}
			------------EPOCH 3---------------
Loss:  tensor(936.7319, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1516.3618, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2358.3413, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(985.0259, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1937.5675, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2659.7769, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1720.9388, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1398.6576, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(877.5627, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(871.4283, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1138.2292, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1081.9054, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(810.3427, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(614.4647, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1198.9028, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1618.1899, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(872.8988, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1624.9946, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1924.1281, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1423.9539, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(952.5553, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1026.6470, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(416.9129, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1278.5386, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1856.0557, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1048.5552, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(562.2258, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1338.1572, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(919.1825, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(768.7223, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1414.1979, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2068.2156, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1808.0612, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1366.3766, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(750.3665, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(940.9360, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(908.8818, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(948.3239, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1346.5618, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1108.4827, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(993.4841, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1217.2983, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1258.6614, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1228.4396, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1100.9622, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1883.4623, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(973.1203, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1208.6223, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1278.9094, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(825.2460, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(478.7328, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(686.9035, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1053.6396, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1365.6353, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1243.5720, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1092.9453, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(972.7729, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(959.7001, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.2413793103448276, 'recall': 0.15498154981549817, 'f1': 0.18876404494382024, 'number': 271}, 'P': {'precision': 0.25970149253731345, 'recall': 0.47540983606557374, 'f1': 0.3359073359073359, 'number': 366}, 'overall_precision': 0.2559241706161137, 'overall_recall': 0.3390894819466248, 'overall_f1': 0.2916948008102633, 'overall_accuracy': 0.6227149946604686}
			------------EPOCH 4---------------
Loss:  tensor(789.6231, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1343.4810, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2106.4570, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(781.8287, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1750.1821, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2204.9375, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1347.0966, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1210.9089, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(709.7284, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(673.8025, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(854.2655, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(807.7068, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(660.1727, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(457.3547, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(884.3406, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1308.2870, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(735.6096, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1293.8588, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1539.2539, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1157.0415, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(772.0061, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(770.9784, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(300.0787, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1023.8813, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1539.5984, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(823.2823, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(443.8046, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(990.2729, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(578.1672, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(618.8659, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(922.4789, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1540.5442, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1348.9707, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1237.5244, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(549.9957, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(763.2131, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(779.1126, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(782.9481, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1083.5671, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(901.5967, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(628.9567, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(521.9144, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(596.5919, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(644.5388, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(924.4114, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1298.2629, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(713.4242, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(885.9266, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(937.4797, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(602.0148, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(336.7072, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(496.6681, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(779.7635, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1105.8600, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(972.0497, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(777.3148, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(766.5363, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(798.2407, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.42011834319526625, 'recall': 0.26199261992619927, 'f1': 0.32272727272727275, 'number': 271}, 'P': {'precision': 0.3256578947368421, 'recall': 0.5409836065573771, 'f1': 0.406570841889117, 'number': 366}, 'overall_precision': 0.3462033462033462, 'overall_recall': 0.42229199372056514, 'overall_f1': 0.38048090523338046, 'overall_accuracy': 0.6936993529744331}
			------------EPOCH 5---------------
Loss:  tensor(683.5582, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(957.6860, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1326.4844, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(642.3395, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1339.1487, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1963.7925, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(938.0380, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(990.6001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(570.0992, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(517.5057, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(644.3895, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(615.1769, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(577.9746, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(393.7427, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(647.1406, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1059.6390, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(546.4615, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(982.1058, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1323.9319, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1045.4491, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(657.8964, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(566.7685, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(224.8885, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(789.3177, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1216.0459, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(658.1150, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(473.9790, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(869.7478, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(369.3702, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(513.6881, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(437.4768, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1317.7180, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1041.7961, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1047.5914, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(448.8401, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(612.8474, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(745.3905, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(625.7852, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(822.7502, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(752.3726, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(469.9563, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(314.6485, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(452.7813, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(477.4437, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(657.1401, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(895.0048, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(512.6431, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(639.3425, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(589.9226, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(385.3036, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(209.0960, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(352.5482, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(634.7683, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1024.5438, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(793.2837, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(673.6432, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(582.5584, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(584.3714, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.39402173913043476, 'recall': 0.5350553505535055, 'f1': 0.4538341158059468, 'number': 271}, 'P': {'precision': 0.436950146627566, 'recall': 0.40710382513661203, 'f1': 0.42149929278642156, 'number': 366}, 'overall_precision': 0.4146685472496474, 'overall_recall': 0.46153846153846156, 'overall_f1': 0.43684992570579495, 'overall_accuracy': 0.710534581317922}
			------------EPOCH 6---------------
Loss:  tensor(445.8304, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(712.9487, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(811.7483, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(443.9465, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(857.1002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1597.8462, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(608.4304, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(535.4855, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(428.3687, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(393.8763, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(478.9242, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(503.9342, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(404.1611, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(357.2926, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(674.1359, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(875.3971, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(499.2210, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1053.1410, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1436.9573, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(934.9615, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(460.7853, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(467.0574, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(193.7960, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(714.0293, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1132.1171, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(558.7662, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(366.3674, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(751.2350, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(563.8729, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(537.8199, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(470.0714, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1424.0374, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(970.7875, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(782.9368, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(480.4819, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(511.0560, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(585.1329, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(548.0733, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(708.0605, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(704.1703, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(418.4822, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(251.7790, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(338.3665, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(384.9189, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(568.4554, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(697.4218, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(528.1508, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(594.9648, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(650.6053, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(340.1215, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(259.8912, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(323.3699, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(652.4564, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(715.0513, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(889.1058, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(587.2649, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(489.7918, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(537.1234, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3783783783783784, 'recall': 0.46494464944649444, 'f1': 0.4172185430463576, 'number': 271}, 'P': {'precision': 0.3957894736842105, 'recall': 0.5136612021857924, 'f1': 0.44708680142687274, 'number': 366}, 'overall_precision': 0.3886138613861386, 'overall_recall': 0.49293563579277866, 'overall_f1': 0.43460207612456747, 'overall_accuracy': 0.6858471009485521}
			------------EPOCH 7---------------
Loss:  tensor(372.8415, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(557.2318, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(908.9019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(357.5078, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(809.3558, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1223.0654, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(577.2136, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(563.6674, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(296.7567, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(360.8306, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(419.6002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(447.9982, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(355.8619, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(252.7332, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(399.3309, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(659.4083, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(471.4104, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(682.6152, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(844.8141, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(550.0995, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(349.4181, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(365.0572, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(123.6211, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(597.9865, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(696.4977, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(550.4869, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(313.1897, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(388.8494, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(227.6057, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(269.5862, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(234.9427, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(932.5959, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(585.6541, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(557.0079, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(240.2576, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(332.1777, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(378.4113, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(359.8131, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(412.2613, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(457.5788, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(324.6405, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(196.0341, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(211.6430, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(269.0114, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(601.5353, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(611.3257, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(380.2798, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(440.9874, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(335.4596, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(183.1093, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(135.1402, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(232.2108, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(359.2426, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(511.1920, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(384.9107, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(242.7965, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(385.6528, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(367.5180, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5306122448979592, 'recall': 0.3837638376383764, 'f1': 0.4453961456102784, 'number': 271}, 'P': {'precision': 0.5272727272727272, 'recall': 0.7131147540983607, 'f1': 0.6062717770034842, 'number': 366}, 'overall_precision': 0.5282199710564399, 'overall_recall': 0.5729984301412873, 'overall_f1': 0.5496987951807228, 'overall_accuracy': 0.7364784220114329}
			------------EPOCH 8---------------
Loss:  tensor(207.5568, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(293.8588, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(479.3914, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(189.0789, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(533.0830, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(978.8270, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(333.0244, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(280.0840, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(174.0513, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(163.8531, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(160.0454, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(208.1518, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(277.6133, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(269.2034, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(241.4063, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(445.1211, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(269.5090, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(444.6201, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(695.6293, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(481.4666, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(231.6429, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(330.6759, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(116.2519, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(431.8374, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(473.2469, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(241.1450, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(221.3427, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(266.8267, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(165.4227, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(202.7109, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(151.4895, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(597.0652, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(360.2183, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(465.2413, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(207.1762, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(248.4647, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(350.7014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(320.3911, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(229.9159, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(333.4138, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(201.8046, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(148.6822, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(168.8960, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(190.1301, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(312.0453, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(440.2724, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(215.4716, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(341.6257, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(266.0831, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(136.0184, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(87.0355, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(177.9156, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(321.9349, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(533.4255, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(279.6177, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(196.6831, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(273.1104, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(345.0656, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.429042904290429, 'recall': 0.4797047970479705, 'f1': 0.45296167247386754, 'number': 271}, 'P': {'precision': 0.5763546798029556, 'recall': 0.639344262295082, 'f1': 0.6062176165803109, 'number': 366}, 'overall_precision': 0.5133991537376587, 'overall_recall': 0.5714285714285714, 'overall_f1': 0.5408618127786033, 'overall_accuracy': 0.748413845090772}
			------------EPOCH 9---------------
Loss:  tensor(108.3778, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(174.3481, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(289.0158, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(136.9566, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(316.6048, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(596.2154, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(171.3210, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(168.1038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(162.6168, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(136.5725, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(136.8696, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(166.0656, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(218.4949, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(180.6045, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(257.0833, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(229.4849, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(218.2396, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(347.4694, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(523.2584, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(268.5413, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(151.3385, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(223.2495, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(72.7181, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(268.8916, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(355.0686, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(168.2873, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(186.8108, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(205.9199, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(160.2812, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(167.7840, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(227.4843, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(545.5414, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(244.1710, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(303.5223, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(153.5065, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(226.8017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(202.4009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(241.8274, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(153.3353, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(274.2928, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(152.2242, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(107.1443, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(116.7531, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(147.2251, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(285.8152, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(395.4080, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(221.1818, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(243.9820, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(173.9519, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(76.5554, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(75.1384, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(113.4465, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(218.7615, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(335.3362, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(192.7342, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(130.0874, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(186.2413, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(239.9782, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.387434554973822, 'recall': 0.5461254612546126, 'f1': 0.4532924961715162, 'number': 271}, 'P': {'precision': 0.6176470588235294, 'recall': 0.4016393442622951, 'f1': 0.48675496688741726, 'number': 366}, 'overall_precision': 0.47580645161290325, 'overall_recall': 0.4631083202511774, 'overall_f1': 0.4693715194908512, 'overall_accuracy': 0.6854701928513097}
			------------EPOCH 10---------------
Loss:  tensor(167.2130, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(399.1875, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(483.7541, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(269.9520, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(284.8139, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(599.8530, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(174.6360, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(188.7365, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(104.7855, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(72.1142, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(79.4749, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(107.8846, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(167.6996, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(138.9180, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(164.4701, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(193.2826, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(179.3772, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(288.9403, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(375.5891, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(248.8331, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(221.5647, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(315.7980, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(88.2682, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(381.6858, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(560.6099, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(474.3819, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(328.2055, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(310.2822, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(293.9313, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(195.4022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(341.3131, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(516.0222, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(271.0677, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(219.2457, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(124.1012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(169.1239, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(249.4938, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(208.9806, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(233.5317, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(303.8543, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(200.2950, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(121.3993, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(105.5937, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(198.3246, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(355.8253, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(305.5575, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(201.7825, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(264.1044, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(236.1762, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(115.6354, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(59.6154, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(151.0520, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(240.5094, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(432.3564, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(248.3487, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(290.7925, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(217.4155, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(161.0718, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5294117647058824, 'recall': 0.33210332103321033, 'f1': 0.40816326530612246, 'number': 271}, 'P': {'precision': 0.5753424657534246, 'recall': 0.6885245901639344, 'f1': 0.626865671641791, 'number': 366}, 'overall_precision': 0.5625, 'overall_recall': 0.5368916797488226, 'overall_f1': 0.5493975903614458, 'overall_accuracy': 0.7271813556127897}
			------------EPOCH 11---------------
Loss:  tensor(105.8277, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(170.5376, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(331.3021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(94.1904, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(282.3448, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(866.3267, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(132.4832, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(126.2337, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(79.0971, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(77.5381, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(75.1965, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(114.6606, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(148.5044, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(116.1431, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(118.4420, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(289.9499, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(140.7625, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(232.4771, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(288.7807, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(174.4111, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(97.5616, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(190.6064, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.1657, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(273.8017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(231.1902, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(94.9995, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(106.5036, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(143.4084, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(140.4317, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(147.4263, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(127.1519, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(290.2904, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(355.2021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(372.3815, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(189.6670, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(246.0826, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(351.5458, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(428.4395, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(247.5242, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(335.1125, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(385.9222, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(579.1260, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(465.6641, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(389.6125, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(313.4850, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(773.4726, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(165.6091, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(164.4986, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(201.7941, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(118.9742, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.7908, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(124.1502, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(267.7416, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(501.1745, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(341.7208, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(221.8821, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(216.5805, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(266.7379, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.39215686274509803, 'recall': 0.36900369003690037, 'f1': 0.38022813688212925, 'number': 271}, 'P': {'precision': 0.4981684981684982, 'recall': 0.37158469945355194, 'f1': 0.42566510172143984, 'number': 366}, 'overall_precision': 0.44696969696969696, 'overall_recall': 0.3704866562009419, 'overall_f1': 0.4051502145922746, 'overall_accuracy': 0.650543375840191}
			------------EPOCH 12---------------
Loss:  tensor(140.8211, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(188.3997, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(353.6937, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(141.9723, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(583.3730, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1102.4939, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(355.2471, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(298.4410, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(177.8335, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(238.5892, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(192.9128, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(198.1911, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(190.5048, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(104.3888, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(174.0001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(290.9595, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(149.3325, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(199.0153, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(271.2067, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(127.7089, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(86.9644, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(193.2322, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.8127, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(143.5237, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(210.9745, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(93.7037, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(129.2148, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(123.0715, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(64.6556, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(85.7747, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(127.8989, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(239.9556, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(249.4982, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(300.7102, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(146.0313, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(156.9973, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(190.4359, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(242.4275, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(243.0655, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(250.3593, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(246.7210, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(131.3374, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(95.0309, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(287.7010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(540.4902, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(919.7557, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(191.8932, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(571.4415, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(357.5063, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(254.3179, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(126.9119, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(184.4599, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(363.9406, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(339.3490, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(341.3917, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(303.4032, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(165.5771, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(110.4041, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.43258426966292135, 'recall': 0.5682656826568265, 'f1': 0.49122807017543857, 'number': 271}, 'P': {'precision': 0.5882352941176471, 'recall': 0.6284153005464481, 'f1': 0.607661822985469, 'number': 366}, 'overall_precision': 0.5140562248995983, 'overall_recall': 0.6028257456828885, 'overall_f1': 0.5549132947976878, 'overall_accuracy': 0.7159997487279351}
			------------EPOCH 13---------------
Loss:  tensor(74.3829, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(150.3365, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(370.0589, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(77.7139, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(224.8390, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(377.8351, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(90.9130, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(107.4490, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(132.4719, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(93.1042, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(117.5021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(122.4812, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(174.9048, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(176.4803, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(223.6568, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(279.9310, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(343.2173, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(352.3091, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(330.3399, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(368.7731, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(186.4898, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(309.7654, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(87.3551, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(242.5717, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(204.4284, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(218.6533, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(319.0111, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(158.0765, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.6703, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(177.5707, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(89.8642, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(334.9476, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(123.6617, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(187.6526, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(126.3376, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(126.3783, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(112.2818, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(146.3409, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(85.4948, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(95.3726, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(81.3786, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(78.1416, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.0405, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(83.9944, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(96.5644, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(120.8534, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(100.2893, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(119.4966, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(141.6872, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(80.8169, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.0454, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(76.2811, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(149.1981, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(287.5312, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(125.9750, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(94.5575, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(127.9775, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(128.1237, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3686746987951807, 'recall': 0.5645756457564576, 'f1': 0.446064139941691, 'number': 271}, 'P': {'precision': 0.5344827586206896, 'recall': 0.5081967213114754, 'f1': 0.5210084033613445, 'number': 366}, 'overall_precision': 0.4442988204456094, 'overall_recall': 0.5321821036106751, 'overall_f1': 0.4842857142857143, 'overall_accuracy': 0.6854701928513097}
			------------EPOCH 14---------------
Loss:  tensor(212.1792, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(173.0874, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(582.8171, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(211.8401, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(479.2415, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(561.2021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(120.8591, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(171.0439, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(80.7566, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(146.9947, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(96.1957, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(107.6966, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(106.9095, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(114.8036, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(110.3023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(167.2317, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(93.5378, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(119.5147, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(133.3342, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(127.3333, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.2539, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(101.3474, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.6236, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(83.9943, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(90.7895, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.0926, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(84.5335, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.2008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.7096, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.1101, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.2921, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(103.0714, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(149.7438, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(171.2431, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(114.4625, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(85.9522, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(105.0424, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(173.5787, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(116.0750, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(149.3162, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(131.0684, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(95.2978, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(105.7317, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(105.2999, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(96.2514, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(126.5709, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(99.8012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(223.9601, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(92.5673, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(71.4755, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.6523, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.3383, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(123.6546, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(166.0970, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(77.6785, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.2297, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.4515, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.7281, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4387254901960784, 'recall': 0.6605166051660517, 'f1': 0.5272459499263624, 'number': 271}, 'P': {'precision': 0.6666666666666666, 'recall': 0.4972677595628415, 'f1': 0.5696400625978091, 'number': 366}, 'overall_precision': 0.5301027900146843, 'overall_recall': 0.5667189952904239, 'overall_f1': 0.5477996965098634, 'overall_accuracy': 0.7124191218041335}
			------------EPOCH 15---------------
Loss:  tensor(29.7626, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.4783, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(91.5877, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.9912, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(107.1111, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(191.8363, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(72.4313, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.9807, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(83.0680, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.5272, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.9122, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(70.4443, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(190.0823, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(150.9563, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(126.0851, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(222.3559, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(82.5250, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(117.4275, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(188.6770, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(129.2936, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.4533, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(109.6773, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.3343, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(70.8998, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(65.7591, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.8548, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.3977, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.3655, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.2539, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.1129, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.9346, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(74.2363, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(112.2255, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(136.2092, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(66.3700, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(72.5843, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(139.4877, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(95.5934, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(114.1188, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(148.3713, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(99.0661, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(87.8289, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(85.3542, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(81.9200, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(78.7494, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(68.7577, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(83.9252, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(110.4763, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(74.5199, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.8672, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.5425, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(74.1738, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(107.2198, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(172.2093, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.0631, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.4757, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(62.3146, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(76.9702, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4421052631578947, 'recall': 0.6199261992619927, 'f1': 0.5161290322580645, 'number': 271}, 'P': {'precision': 0.6130952380952381, 'recall': 0.5628415300546448, 'f1': 0.5868945868945868, 'number': 366}, 'overall_precision': 0.5223463687150838, 'overall_recall': 0.5871271585557299, 'overall_f1': 0.5528455284552846, 'overall_accuracy': 0.7307619825365915}
			------------EPOCH 16---------------
Loss:  tensor(19.5741, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.9506, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(77.8414, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.0065, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(81.7255, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(166.6192, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.1793, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.0348, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.7392, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.3628, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.3968, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.4558, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.1747, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.9724, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(62.6973, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.5983, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(133.0476, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(141.7982, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(127.6227, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(99.8573, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(338.9856, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(405.3176, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(221.1620, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(739.8528, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(185.4245, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(96.7010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(125.2856, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(144.9005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.5812, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.6304, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.2443, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(70.7356, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(89.7296, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.5195, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.9982, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.6018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.3868, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.1456, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.0595, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(76.0075, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(91.7721, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(67.0476, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(73.3654, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(108.0038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(474.7419, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(350.2641, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(342.9172, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(401.0377, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(288.0374, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.9160, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(89.2040, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(139.9212, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(200.9012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(169.4626, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(81.3963, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(109.0493, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.7069, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(73.1325, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4781144781144781, 'recall': 0.5239852398523985, 'f1': 0.5, 'number': 271}, 'P': {'precision': 0.5941320293398533, 'recall': 0.6639344262295082, 'f1': 0.6270967741935484, 'number': 366}, 'overall_precision': 0.5453257790368272, 'overall_recall': 0.6043956043956044, 'overall_f1': 0.573343261355175, 'overall_accuracy': 0.7425089515673094}
			------------EPOCH 17---------------
Loss:  tensor(14.2646, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.2444, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(64.3099, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.6123, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(73.4988, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(118.9639, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.5347, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.0519, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.3737, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.8305, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.1719, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.1268, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.8283, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.5075, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(62.8294, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(85.4214, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(76.7968, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(92.0812, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(132.2621, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(103.4822, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(94.2841, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(156.2444, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.7397, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(180.1618, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(241.7819, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(290.2744, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(362.3502, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(916.1339, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(406.9559, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(314.4517, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(131.3765, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(800.8629, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(120.8528, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(80.2169, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(89.7431, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(105.6575, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.7765, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.6077, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.1888, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.5104, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.1637, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.8826, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.0108, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(59.2535, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(87.6557, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(75.1889, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(77.8132, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(67.5953, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(311.7164, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(125.3374, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(84.6629, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(144.1514, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1009.9460, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1192.9102, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1988.2881, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1007.0303, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(726.2195, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1039.8676, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5652173913043478, 'recall': 0.23985239852398524, 'f1': 0.33678756476683935, 'number': 271}, 'P': {'precision': 0.49322033898305084, 'recall': 0.7950819672131147, 'f1': 0.6087866108786611, 'number': 366}, 'overall_precision': 0.5049645390070922, 'overall_recall': 0.5588697017268446, 'overall_f1': 0.5305514157973173, 'overall_accuracy': 0.7030592373892832}
			------------EPOCH 18---------------
Loss:  tensor(72.7635, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(333.5338, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(528.6834, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(184.8904, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(303.6279, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(549.6260, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(131.5231, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(234.1188, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(71.8421, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.3874, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.4528, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(89.0525, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.2825, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.2667, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(64.4845, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.0622, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(66.7645, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(82.1702, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(100.8300, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(66.4880, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(123.5826, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(123.4153, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.8633, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(138.5565, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(434.6061, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(112.8291, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(104.3502, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(219.5898, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(695.8740, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(199.5882, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(230.3427, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1002.1749, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1601.7726, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(582.3280, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(472.0165, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(289.2414, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(245.1823, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(216.5829, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1083.0857, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(423.9673, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(258.1746, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(102.5836, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(117.2405, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(121.8149, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(113.4448, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(191.0361, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(83.1726, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(86.2799, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(101.3576, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(59.6659, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.7373, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.0163, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(179.6887, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(244.9723, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(148.6894, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(102.3618, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(194.6056, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(158.1225, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.625, 'recall': 0.16605166051660517, 'f1': 0.26239067055393583, 'number': 271}, 'P': {'precision': 0.499194847020934, 'recall': 0.8469945355191257, 'f1': 0.6281661600810537, 'number': 366}, 'overall_precision': 0.5122655122655123, 'overall_recall': 0.5572998430141287, 'overall_f1': 0.5338345864661653, 'overall_accuracy': 0.7029964193730762}
			------------EPOCH 19---------------
Loss:  tensor(86.2176, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(385.0137, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(568.0994, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(234.0441, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(760.7977, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(972.5687, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(410.2372, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(436.5451, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(475.4200, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(218.5194, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(212.7411, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(397.6972, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(304.8164, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(171.2859, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(350.4597, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(199.1735, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(127.9540, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(304.1443, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(547.3201, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(215.7359, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(59.2623, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(165.4636, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.2381, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(353.0487, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(94.9789, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.8711, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(61.3858, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(69.0699, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.1354, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.9940, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.5866, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(134.7870, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(142.7799, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(117.4505, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.0683, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(86.1436, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(82.8590, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(136.0767, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(207.9086, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(171.4730, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(169.5983, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(72.4019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(111.3590, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(170.1211, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(758.3752, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(519.3826, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(287.9766, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(505.2500, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(444.8298, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(139.0580, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(78.8512, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(233.6850, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(414.4725, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(800.0350, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(270.3563, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(580.0710, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(214.6091, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(202.3755, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4088541666666667, 'recall': 0.5793357933579336, 'f1': 0.4793893129770993, 'number': 271}, 'P': {'precision': 0.6636085626911316, 'recall': 0.592896174863388, 'f1': 0.6262626262626263, 'number': 366}, 'overall_precision': 0.5260196905766527, 'overall_recall': 0.5871271585557299, 'overall_f1': 0.5548961424332345, 'overall_accuracy': 0.7260506313210628}
			------------EPOCH 20---------------
Loss:  tensor(30.6822, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(73.5009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(80.5765, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.9881, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(79.4820, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(178.1955, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.5217, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.5109, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.6476, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.2194, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.7280, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.1598, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.0805, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.4098, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(79.7714, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.8545, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(83.8859, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(95.6478, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(172.9733, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(78.1044, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.3020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(196.2476, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.3958, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(286.3584, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(285.4796, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(220.6229, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(140.3812, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(150.2410, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(96.8717, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(77.5293, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(242.1917, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(994.6639, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(232.8398, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(237.5666, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(127.9549, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(145.5912, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(78.3850, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(66.0841, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(64.1446, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(71.5868, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.5324, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.2140, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.3568, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.2997, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.3560, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.5457, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.2430, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(59.1856, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.8491, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.7773, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.7432, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.2429, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(90.2479, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(145.2959, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.6585, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.9829, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.8496, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.8788, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3923076923076923, 'recall': 0.5645756457564576, 'f1': 0.46293494704992433, 'number': 271}, 'P': {'precision': 0.6403785488958991, 'recall': 0.5546448087431693, 'f1': 0.5944363103953147, 'number': 366}, 'overall_precision': 0.5035360678925035, 'overall_recall': 0.5588697017268446, 'overall_f1': 0.5297619047619047, 'overall_accuracy': 0.711916577674477}
			------------EPOCH 21---------------
Loss:  tensor(15.2304, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.6793, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(66.9742, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.3116, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(77.9281, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(126.8913, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.4119, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.5638, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.4761, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.4866, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.3450, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.0379, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.0791, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.0124, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.9805, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(204.2387, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.3824, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.2182, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(79.3654, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.8649, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.9901, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(72.6520, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.6672, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(83.4830, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.9874, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.3654, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.0713, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.5337, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.5066, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.8281, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.7620, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(99.8645, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(90.4297, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.1324, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.5879, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.3355, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.7903, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.7483, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.5452, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.4164, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.8968, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.2428, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.2183, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.6984, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.0604, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.8081, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.9630, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.5733, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.0091, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.7316, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.5175, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.8035, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(80.6301, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(135.9444, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.9702, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.3412, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.3094, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.9421, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3907766990291262, 'recall': 0.5940959409594095, 'f1': 0.47144948755490484, 'number': 271}, 'P': {'precision': 0.6123778501628665, 'recall': 0.5136612021857924, 'f1': 0.5586924219910848, 'number': 366}, 'overall_precision': 0.4853963838664812, 'overall_recall': 0.5478806907378336, 'overall_f1': 0.5147492625368731, 'overall_accuracy': 0.6908097242289089}
			------------EPOCH 22---------------
Loss:  tensor(11.6252, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.6643, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.6120, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.9096, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(64.0578, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(81.4304, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.8728, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.1548, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.1268, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.5771, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.5715, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.7893, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.1984, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.7835, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.1418, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(111.2375, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.9861, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.7497, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(59.7104, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.7523, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.0996, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.2539, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.3277, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.6842, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.4265, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.7500, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.4751, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.2035, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.0212, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.1564, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.5764, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(59.8549, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(77.6917, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.7071, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.4371, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.2245, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.6529, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.3093, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.1575, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.1634, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.2875, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.1593, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.3936, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.7940, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.0568, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.4498, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.7038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.1730, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.3003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.4516, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.4897, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.3133, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.8688, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(111.0594, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.4172, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.6407, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.4587, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.7218, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4277286135693215, 'recall': 0.5350553505535055, 'f1': 0.47540983606557374, 'number': 271}, 'P': {'precision': 0.6310160427807486, 'recall': 0.644808743169399, 'f1': 0.6378378378378378, 'number': 366}, 'overall_precision': 0.5343618513323983, 'overall_recall': 0.598116169544741, 'overall_f1': 0.5644444444444444, 'overall_accuracy': 0.7372322382059174}
			------------EPOCH 23---------------
Loss:  tensor(7.3224, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.4705, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.7633, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.7347, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.8224, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(62.0821, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.9260, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.3052, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.7943, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.8986, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.6896, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.6677, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.9635, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.8055, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.4074, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.1957, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.6315, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.5468, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.4832, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.5277, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.9991, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.2688, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.4966, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.7336, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.0768, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.0736, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.9973, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.6970, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.3912, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.0766, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.9965, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.8194, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(73.0303, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.0062, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.9148, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.7509, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.7402, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.2080, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.2253, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.6291, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.2505, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.2030, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.0022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.5562, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.1115, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.9040, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.8566, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.7537, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.2362, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.4293, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.9842, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.1864, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.6440, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(96.8552, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.5717, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.0757, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.8504, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.2878, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.42042042042042044, 'recall': 0.5166051660516605, 'f1': 0.4635761589403974, 'number': 271}, 'P': {'precision': 0.6289473684210526, 'recall': 0.6530054644808743, 'f1': 0.6407506702412867, 'number': 366}, 'overall_precision': 0.5315568022440392, 'overall_recall': 0.5949764521193093, 'overall_f1': 0.5614814814814815, 'overall_accuracy': 0.7369181481248822}
			------------EPOCH 24---------------
Loss:  tensor(5.8023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.1696, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.7697, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.0763, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.2481, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.3096, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.1390, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.5023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.8137, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.1454, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.6353, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.4662, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.5277, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.8332, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.5155, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.1149, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.3098, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.4839, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.8896, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.4143, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.6259, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.1012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.6277, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.9245, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.9243, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.9616, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.8449, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.8038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.2413, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.6913, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.0019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.3354, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(71.2522, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.4061, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.6639, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.3024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.7286, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.7557, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.4993, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.3993, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.5000, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.3303, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.2997, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.7299, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.2909, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.3073, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.2085, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.1713, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.2233, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.6134, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.6791, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.5355, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.2933, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(86.1363, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.8042, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.8715, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.2144, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.3676, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4199395770392749, 'recall': 0.5129151291512916, 'f1': 0.46179401993355484, 'number': 271}, 'P': {'precision': 0.6204188481675392, 'recall': 0.6475409836065574, 'f1': 0.6336898395721926, 'number': 366}, 'overall_precision': 0.5273492286115007, 'overall_recall': 0.5902668759811617, 'overall_f1': 0.557037037037037, 'overall_accuracy': 0.7335259752497016}
			------------EPOCH 25---------------
Loss:  tensor(5.0090, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.4621, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.1141, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.4756, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.8094, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.3738, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.1657, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.1475, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.3139, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.6083, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.4256, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.2827, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.6224, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.9079, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.7864, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.7289, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.3811, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.7836, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.3449, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.0944, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.8123, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.8338, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.0523, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.8492, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.4063, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.6881, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.7545, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.0292, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.1390, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.7806, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.7857, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.5558, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(69.3476, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.3772, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.3723, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.2420, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.2634, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.7013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.3796, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.3301, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.6173, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.3773, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.6690, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.7953, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.2333, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.8571, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.5941, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.0697, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.8841, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.8971, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.3750, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.8162, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.0349, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(75.5261, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.9029, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.5405, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.9143, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.4234, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.41369047619047616, 'recall': 0.5129151291512916, 'f1': 0.4579901153212521, 'number': 271}, 'P': {'precision': 0.6327077747989276, 'recall': 0.644808743169399, 'f1': 0.6387009472259811, 'number': 366}, 'overall_precision': 0.5289139633286318, 'overall_recall': 0.5886970172684458, 'overall_f1': 0.5572065378900445, 'overall_accuracy': 0.7339028833469439}
			------------EPOCH 26---------------
Loss:  tensor(4.5089, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.6258, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.6358, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.1672, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.0101, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.4269, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.0483, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.3974, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.1446, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.7001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.5409, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.9520, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.8441, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.0890, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.4270, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.1739, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.6467, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.2235, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.7133, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.4660, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.4293, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.1921, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.1909, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.4581, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.5144, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.9175, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.9994, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.8143, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.4635, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.1981, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.6946, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.0831, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(66.3874, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.4469, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.2601, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.9229, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.0700, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.1567, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.3390, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.5147, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.6513, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.6358, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.8947, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.9409, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.9485, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.0098, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.5877, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.9775, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.4931, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.1924, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.3541, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.8564, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.3589, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(66.8240, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.4689, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.0747, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.7498, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.0588, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3935860058309038, 'recall': 0.4981549815498155, 'f1': 0.4397394136807818, 'number': 271}, 'P': {'precision': 0.6307277628032345, 'recall': 0.639344262295082, 'f1': 0.6350067842605156, 'number': 366}, 'overall_precision': 0.5168067226890757, 'overall_recall': 0.5792778649921507, 'overall_f1': 0.5462620281273131, 'overall_accuracy': 0.7295684402286576}
			------------EPOCH 27---------------
Loss:  tensor(3.8527, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.4931, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.3130, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.1512, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.4115, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.0199, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.1242, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.5222, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.2918, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.9437, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.6519, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.7847, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.1258, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.8319, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.3857, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.5451, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.9438, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.8265, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.3283, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.2922, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.3274, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.7003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.6216, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.5235, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.7968, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.3890, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.6445, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.0505, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.9341, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.7820, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.9009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.7757, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(68.9221, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.4044, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.0656, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.3409, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.2193, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.9627, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.0062, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.7182, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.4867, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.9146, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.8341, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.0789, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.3221, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.9159, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.9653, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.7791, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.6938, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.7839, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.8371, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.0385, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.6692, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.3801, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.9047, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.0952, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.8071, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.1820, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.38571428571428573, 'recall': 0.4981549815498155, 'f1': 0.43478260869565216, 'number': 271}, 'P': {'precision': 0.6243243243243243, 'recall': 0.6311475409836066, 'f1': 0.6277173913043478, 'number': 366}, 'overall_precision': 0.5083333333333333, 'overall_recall': 0.5745682888540031, 'overall_f1': 0.5394252026529107, 'overall_accuracy': 0.7181355612789748}
			------------EPOCH 28---------------
Loss:  tensor(3.6192, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.9641, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.0117, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.9827, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.1696, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.0327, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.9540, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.8005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.4255, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.2506, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.2166, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.8033, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.5167, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.8117, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.4126, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.8873, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.9463, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.2998, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.5823, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.8824, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.1184, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.6603, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.3114, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.5471, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.2066, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.6291, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.4912, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.2832, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.6293, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.9524, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.2853, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.9949, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.8474, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.5062, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.2764, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.6261, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.4799, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.5536, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.0601, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.4867, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.9654, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.7830, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.9683, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.6221, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.4477, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.0451, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.5859, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.8235, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.5931, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.8970, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.7898, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.1108, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.1227, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.3917, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.5978, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.7081, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.9404, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.5076, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.41040462427745666, 'recall': 0.5239852398523985, 'f1': 0.460291734197731, 'number': 271}, 'P': {'precision': 0.623342175066313, 'recall': 0.6420765027322405, 'f1': 0.63257065948856, 'number': 366}, 'overall_precision': 0.5214384508990318, 'overall_recall': 0.5918367346938775, 'overall_f1': 0.5544117647058824, 'overall_accuracy': 0.7188893774734594}
			------------EPOCH 29---------------
Loss:  tensor(5.7377, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.5356, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.0714, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.3919, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.6088, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.2482, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.8812, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.4606, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.1526, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.9788, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.9586, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.7075, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.6783, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.5852, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.7911, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.9028, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.3470, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.9367, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.9759, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.3703, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.0883, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.8820, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.7598, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.7694, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.9217, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.1204, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.2130, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.3741, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.7144, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.9972, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.2942, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.0348, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(67.1797, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.3922, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.0320, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.1598, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.0941, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.7044, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.9754, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.0956, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.4640, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.3377, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.5584, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.6474, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.3271, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.2081, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.1209, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.8777, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.3540, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.9971, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.5019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.6604, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.9524, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.2421, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.8499, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.7655, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.5509, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.5266, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.41543026706231456, 'recall': 0.5166051660516605, 'f1': 0.46052631578947373, 'number': 271}, 'P': {'precision': 0.6200527704485488, 'recall': 0.6420765027322405, 'f1': 0.6308724832214765, 'number': 366}, 'overall_precision': 0.5237430167597765, 'overall_recall': 0.5886970172684458, 'overall_f1': 0.5543237250554324, 'overall_accuracy': 0.7188265594572524}
			------------EPOCH 30---------------
Loss:  tensor(4.7792, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.0387, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.7947, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.7096, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.5554, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.9649, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.6141, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.7821, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.3402, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.9819, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.9946, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.4373, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.9979, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.3796, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.4480, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.4828, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.4689, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.5376, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.2600, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.0063, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.3615, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.4666, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.0814, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.1074, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.5005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.9118, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.7885, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.0209, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.5211, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.8173, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.2689, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.4713, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.6531, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.2877, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.6144, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.0938, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.3534, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.4032, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.7355, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.6200, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.1422, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.6132, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.6925, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.7794, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.7618, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.7642, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.7812, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.2716, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.4809, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.3160, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.8904, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.2158, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.5205, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.1119, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.3191, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.4892, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.4755, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.5546, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4229607250755287, 'recall': 0.5166051660516605, 'f1': 0.46511627906976744, 'number': 271}, 'P': {'precision': 0.6306068601583114, 'recall': 0.6530054644808743, 'f1': 0.6416107382550336, 'number': 366}, 'overall_precision': 0.5338028169014084, 'overall_recall': 0.5949764521193093, 'overall_f1': 0.562731997030438, 'overall_accuracy': 0.7202085558138074}


		-------------RUN 3-----------
			------------EPOCH 1---------------
Loss:  tensor(830.8151, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1512.2786, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3370.5615, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1629.9454, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1813.5957, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1960.3634, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1827.2677, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3527.3206, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3365.8779, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3292.2290, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2030.5243, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1011.9066, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1932.4225, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1651.1042, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1401.9941, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(913.4284, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3863.7131, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2613.9133, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2357.3914, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1675.9756, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1159.6044, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2064.7573, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2331.3403, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2345.0498, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2134.8914, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1209.3877, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1416.3689, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1985.0165, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(921.2334, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1355.1727, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2063.3242, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2570.2329, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2661.0142, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2456.0627, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1769.8636, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1792.0775, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1272.8403, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1835.3828, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2769.0420, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1603.9661, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1672.5339, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1349.7529, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1964.6011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2840.8384, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1084.8346, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1899.1069, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1986.1689, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1550.6140, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1315.5122, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2429.1787, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1690.3357, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1418.0376, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2014.0234, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1751.6102, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1903.7421, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2607.0071, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.034026465028355386, 'recall': 0.05642633228840126, 'f1': 0.04245283018867924, 'number': 319}, 'P': {'precision': 0.08163265306122448, 'recall': 0.09592326139088729, 'f1': 0.08820286659316427, 'number': 417}, 'overall_precision': 0.05691854759568204, 'overall_recall': 0.07880434782608696, 'overall_f1': 0.06609686609686609, 'overall_accuracy': 0.5630006076340938}
			------------EPOCH 2---------------
Loss:  tensor(502.9149, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(901.8138, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2009.7399, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(982.0729, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1226.7485, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1276.4554, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1287.1360, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2367.3442, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2341.6313, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2377.5552, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1353.7749, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(718.2625, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1424.3239, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1235.3036, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1048.2993, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(721.2939, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3007.0996, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2003.0292, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1842.6232, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1493.7417, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(977.6152, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1795.7164, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1719.4608, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1896.4944, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1727.0396, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(981.4545, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1065.8513, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1544.1167, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(788.9944, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1149.6200, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1603.8840, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1995.5267, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2051.9341, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1821.9646, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1368.0245, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1473.2656, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1038.2734, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1515.4767, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2265.9697, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1383.4910, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1410.2958, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1057.9792, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1613.5598, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2235.9297, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(798.3905, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1366.8599, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1773.2736, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1122.4414, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1001.2547, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1843.2871, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1243.9014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1046.8275, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1478.6517, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1333.7365, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1531.8169, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1967.7000, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.06382978723404255, 'recall': 0.11285266457680251, 'f1': 0.08154020385050961, 'number': 319}, 'P': {'precision': 0.22419354838709676, 'recall': 0.3333333333333333, 'f1': 0.26808100289296044, 'number': 417}, 'overall_precision': 0.14780405405405406, 'overall_recall': 0.23777173913043478, 'overall_f1': 0.18229166666666666, 'overall_accuracy': 0.5916146495056068}
			------------EPOCH 3---------------
Loss:  tensor(356.1675, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(706.3989, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1534.1843, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(761.1493, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1039.1736, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(999.0270, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1078.0555, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2027.7725, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1938.7261, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1969.3746, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1130.2375, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(590.2133, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1109.8909, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(974.8052, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(882.2239, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(593.5939, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2434.9214, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1511.7703, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1329.3340, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1032.5116, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(658.1710, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1229.0933, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1404.0466, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1566.2075, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1394.5167, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(763.3916, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(844.4356, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1225.2180, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(638.8037, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(930.0725, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1242.0088, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1446.3431, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1565.7379, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1421.5649, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1061.8905, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1132.6278, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(822.1040, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1117.3618, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1603.4165, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1115.2424, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1227.3398, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(838.8275, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1332.7628, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1696.0521, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(591.6993, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(945.6000, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1438.7729, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(806.9464, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(818.2538, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1375.7686, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(945.2236, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(656.1816, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1182.5896, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(946.8394, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1198.9980, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1380.3638, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.2398753894080997, 'recall': 0.2413793103448276, 'f1': 0.240625, 'number': 319}, 'P': {'precision': 0.34065934065934067, 'recall': 0.4460431654676259, 'f1': 0.3862928348909657, 'number': 417}, 'overall_precision': 0.3033448673587082, 'overall_recall': 0.35733695652173914, 'overall_f1': 0.32813474734872117, 'overall_accuracy': 0.6658564878749379}
			------------EPOCH 4---------------
Loss:  tensor(252.5453, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(596.9634, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1279.8685, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(626.6843, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(788.1741, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(822.0591, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(771.4430, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1239.8484, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1329.3901, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1452.9879, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(895.3592, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(455.9715, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(813.5368, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(740.8699, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(667.4670, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(469.8673, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1872.9297, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1099.4150, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(871.5854, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(463.9359, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(390.1447, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(691.6815, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1018.0724, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1151.7915, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1103.2089, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(584.8735, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(555.4786, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(925.3750, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(545.1196, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(675.6971, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(904.8887, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1006.0206, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1170.2247, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1186.2136, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(848.2076, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(928.4618, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(670.0696, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(801.5707, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1133.9330, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(788.5869, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(992.2567, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(613.2520, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(974.9436, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1160.8782, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(398.5937, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(598.9506, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1051.9165, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(528.8963, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(550.7745, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(991.7439, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(660.7727, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(420.0054, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(864.8402, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(523.5765, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1033.6859, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(909.2058, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.31462925851703405, 'recall': 0.49216300940438873, 'f1': 0.38386308068459657, 'number': 319}, 'P': {'precision': 0.5782312925170068, 'recall': 0.2038369304556355, 'f1': 0.30141843971631205, 'number': 417}, 'overall_precision': 0.3746130030959752, 'overall_recall': 0.328804347826087, 'overall_f1': 0.3502170767004341, 'overall_accuracy': 0.5731646688394189}
			------------EPOCH 5---------------
Loss:  tensor(319.3898, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(624.9603, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1197.6470, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(715.6016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(628.2217, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(554.4269, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(831.5272, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(946.4429, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(764.7283, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1006.9189, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(637.9106, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(302.2789, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(644.4443, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(504.4318, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(459.0894, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(332.2156, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1806.8217, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(955.3033, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(820.3865, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(499.1172, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(322.7686, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(606.8755, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(817.8599, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(865.8974, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(936.1885, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(439.8918, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(469.9500, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(752.3087, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(462.5869, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(682.0995, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(930.9652, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1030.6902, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(997.2951, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(993.6447, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(685.4779, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(759.2717, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(618.1883, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(652.9653, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1013.9578, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(748.6028, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(845.3655, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(804.7494, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1030.7921, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1276.3533, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(524.2182, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(726.6552, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(875.4480, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(640.8105, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(553.1378, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(998.8064, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(654.8882, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(468.8056, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1050.2349, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(782.3563, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1077.3528, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(992.7715, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.34782608695652173, 'recall': 0.32601880877742945, 'f1': 0.3365695792880259, 'number': 319}, 'P': {'precision': 0.35093167701863354, 'recall': 0.5419664268585132, 'f1': 0.4260131950989633, 'number': 417}, 'overall_precision': 0.34994697773064687, 'overall_recall': 0.4483695652173913, 'overall_f1': 0.3930911256700417, 'overall_accuracy': 0.6356957410373971}
			------------EPOCH 6---------------
Loss:  tensor(247.2611, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(556.2657, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1228.6880, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(616.4111, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(906.0769, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(841.9907, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1027.2793, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1465.9594, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1185.8398, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1267.8322, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(819.6055, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(469.3528, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(687.4586, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(558.5880, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(580.6023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(390.7334, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1216.2598, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(682.5718, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(514.4480, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(325.9808, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(267.0439, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(365.0945, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(694.9520, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(677.0343, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(677.3441, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(409.4813, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(524.3956, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(597.8673, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(399.8639, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(463.5479, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(734.3540, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(833.1161, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(858.8496, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(856.4984, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(631.7411, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(642.7567, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(461.0502, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(523.8700, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(678.8721, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(614.1794, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(641.9457, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(410.2188, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(705.0851, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(793.0112, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(281.7153, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(454.9904, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(665.3226, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(450.4452, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(412.0150, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(578.8647, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(403.9695, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(239.7136, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(638.2566, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(284.4109, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(531.7125, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(498.3536, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3554006968641115, 'recall': 0.31974921630094044, 'f1': 0.3366336633663366, 'number': 319}, 'P': {'precision': 0.40214067278287463, 'recall': 0.6306954436450839, 'f1': 0.4911297852474323, 'number': 417}, 'overall_precision': 0.38788522848034007, 'overall_recall': 0.49592391304347827, 'overall_f1': 0.4353011329755516, 'overall_accuracy': 0.6793349168646081}
			------------EPOCH 7---------------
Loss:  tensor(182.2021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(479.9623, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(910.1229, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(355.6796, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(608.9559, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(639.8340, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(598.1446, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(541.5000, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(583.2782, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(639.7664, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(412.1271, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(263.7241, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(465.4318, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(246.7081, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(208.7135, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(217.3616, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(977.7831, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(370.7066, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(427.5217, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(246.8846, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(185.2217, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(240.5096, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(657.3143, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(586.5273, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(436.4624, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(270.9133, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(312.5276, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(372.4606, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(265.0123, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(298.9880, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(417.7991, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(480.6786, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(500.8981, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(703.4047, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(501.7260, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(430.0851, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(258.2854, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(256.9828, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(423.8178, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(322.0776, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(392.4619, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(268.6552, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(399.9012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(434.8828, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(183.6222, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(329.1470, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(476.1274, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(288.4427, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(400.9302, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(469.0222, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(425.1008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(198.4744, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(574.5658, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(210.4189, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(367.1850, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(309.4120, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4420731707317073, 'recall': 0.45454545454545453, 'f1': 0.4482225656877898, 'number': 319}, 'P': {'precision': 0.538899430740038, 'recall': 0.6810551558752997, 'f1': 0.6016949152542374, 'number': 417}, 'overall_precision': 0.5017543859649123, 'overall_recall': 0.5828804347826086, 'overall_f1': 0.5392834695160276, 'overall_accuracy': 0.7237474451748329}
			------------EPOCH 8---------------
Loss:  tensor(133.8694, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(341.1282, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(573.9490, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(209.6693, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(276.1170, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(295.4355, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(327.2791, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(230.6619, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(321.8853, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(361.9193, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(262.2265, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(166.2609, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(284.5532, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(153.9989, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(148.1605, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(148.1690, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(677.2410, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(260.0289, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(278.0535, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(180.2718, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(136.7743, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(193.1943, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(438.3259, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(316.3356, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(469.6017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(227.7589, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(251.7804, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(308.0811, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(280.6617, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(271.8987, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(333.7654, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(425.2902, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(331.0501, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(487.5890, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(401.1742, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(392.4502, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(158.5419, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(175.7278, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(374.5935, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(224.5166, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(326.4679, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(214.4149, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(270.7608, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(306.6700, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(117.8990, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(207.5972, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(313.4427, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(160.7303, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(215.1126, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(316.5239, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(272.9865, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(129.3279, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(644.9902, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(149.4172, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(280.6514, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(242.8426, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.35855263157894735, 'recall': 0.34169278996865204, 'f1': 0.34991974317817015, 'number': 319}, 'P': {'precision': 0.5672514619883041, 'recall': 0.46522781774580335, 'f1': 0.5111989459815547, 'number': 417}, 'overall_precision': 0.46904024767801855, 'overall_recall': 0.4116847826086957, 'overall_f1': 0.4384949348769899, 'overall_accuracy': 0.6644755012981274}
			------------EPOCH 9---------------
Loss:  tensor(83.7996, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(256.2349, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(467.4277, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(225.4927, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(284.6465, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(216.9890, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(494.2848, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(245.6215, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(436.6133, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(357.1975, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(251.8799, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(125.6523, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(188.9635, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(87.8602, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(95.4918, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(86.2422, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(403.3106, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(126.6167, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(170.9392, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(129.8724, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(141.5705, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(185.1131, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(400.6553, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(293.7915, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(284.4482, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(189.7911, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(183.8466, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(188.2960, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(221.4724, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(258.8415, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(260.4256, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(376.9535, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(508.7355, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(421.1520, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(426.3264, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(351.6265, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(225.9692, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(574.5917, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1484.7375, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(344.3344, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(375.2734, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(185.6342, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(325.1092, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(442.7487, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(84.4799, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(129.3666, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(311.5445, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(106.1339, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(178.8822, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(262.1341, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(219.9966, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(104.0367, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(676.1536, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(215.4158, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(294.0963, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(350.8644, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3617021276595745, 'recall': 0.10658307210031348, 'f1': 0.16464891041162227, 'number': 319}, 'P': {'precision': 0.48705882352941177, 'recall': 0.49640287769784175, 'f1': 0.491686460807601, 'number': 417}, 'overall_precision': 0.464354527938343, 'overall_recall': 0.327445652173913, 'overall_f1': 0.3840637450199203, 'overall_accuracy': 0.6116113351378225}
			------------EPOCH 10---------------
Loss:  tensor(110.6367, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(553.9478, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1186.0822, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(602.2607, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(506.6409, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(642.4260, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1081.8147, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(739.3452, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(319.5016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(441.1584, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(445.1666, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(164.4099, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(247.6053, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(199.0840, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(152.1812, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(138.9118, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1252.4648, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(395.4601, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(354.2970, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(177.9585, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(98.8493, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(161.7956, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(469.9040, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(343.1432, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(353.5881, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(179.4399, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(202.6628, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(292.9025, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(327.2298, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(337.4507, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(339.4964, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(468.3176, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(759.5089, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(487.7816, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(487.5138, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(540.9023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(232.4217, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(425.1312, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1096.0796, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(636.3837, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(916.3937, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(311.9980, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(638.0469, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(926.1388, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(147.5121, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(253.8645, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1435.3152, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(324.2842, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(414.1891, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(562.3020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(621.5781, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(363.8092, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(473.6682, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(411.1428, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(706.7999, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(345.8816, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3070539419087137, 'recall': 0.23197492163009403, 'f1': 0.26428571428571423, 'number': 319}, 'P': {'precision': 0.4884437596302003, 'recall': 0.7601918465227818, 'f1': 0.5947467166979362, 'number': 417}, 'overall_precision': 0.4393258426966292, 'overall_recall': 0.53125, 'overall_f1': 0.48093480934809346, 'overall_accuracy': 0.6811025796829255}
			------------EPOCH 11---------------
Loss:  tensor(83.4906, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(221.7879, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(343.9305, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(130.4027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(289.3301, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(391.0179, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(547.1870, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(355.3538, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(256.0533, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(383.1194, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(458.8022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(268.1423, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(199.0844, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(130.0097, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(109.0424, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(130.2096, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(440.5442, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(281.9185, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(194.8492, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(152.5593, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(177.0372, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(194.5678, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(707.9142, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(552.4111, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(487.1971, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(390.5257, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(276.7630, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(601.1991, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(138.5506, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(163.8015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(240.8611, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(265.0412, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(215.1532, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(389.1381, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(255.4343, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(206.5838, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(192.9401, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(179.9310, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(338.3995, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(228.0364, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(233.5446, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(266.8293, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(322.1208, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(286.0023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(118.1307, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(166.3147, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(356.7737, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(138.2587, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(170.9049, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(232.4376, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(235.1637, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(95.1386, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(236.3013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(76.0612, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(197.6198, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(106.1068, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4438040345821326, 'recall': 0.4827586206896552, 'f1': 0.4624624624624625, 'number': 319}, 'P': {'precision': 0.49705304518664045, 'recall': 0.6067146282973621, 'f1': 0.5464362850971922, 'number': 417}, 'overall_precision': 0.47546728971962615, 'overall_recall': 0.5529891304347826, 'overall_f1': 0.5113065326633165, 'overall_accuracy': 0.7036402806164724}
			------------EPOCH 12---------------
Loss:  tensor(62.5332, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(183.6989, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(245.1147, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(119.7179, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(123.7820, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(159.3457, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(204.6370, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(157.3947, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(143.5265, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(204.8987, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(85.5979, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.4313, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(135.0908, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(72.8507, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(70.5332, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.0180, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(265.9070, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(108.1864, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(114.1363, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(125.5027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(78.6618, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(109.4094, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(246.7108, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(118.1005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(143.3605, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(99.7168, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(84.5391, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(94.7661, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.1711, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(93.5017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(128.0668, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(115.5627, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(164.1304, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(217.9715, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(158.6174, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(153.4874, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(98.6165, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(145.9121, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(179.6495, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(146.1692, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(150.4311, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(123.3248, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(239.0839, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(185.9349, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.4882, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(94.2621, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(173.3232, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(67.4352, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(90.3291, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(122.3574, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(148.1726, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.5847, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(137.9048, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.9151, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(116.9509, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(89.3067, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5681818181818182, 'recall': 0.31347962382445144, 'f1': 0.4040404040404041, 'number': 319}, 'P': {'precision': 0.4588938714499253, 'recall': 0.7362110311750599, 'f1': 0.5653775322283611, 'number': 417}, 'overall_precision': 0.4816568047337278, 'overall_recall': 0.5529891304347826, 'overall_f1': 0.514864010120177, 'overall_accuracy': 0.7036955200795448}
			------------EPOCH 13---------------
Loss:  tensor(32.7917, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(124.3182, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(190.0437, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(106.4555, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(127.8107, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(209.2298, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(169.5738, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(110.7152, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(143.0777, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(217.0612, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(119.6677, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(91.3249, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(108.5527, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(86.8089, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(102.5623, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(83.9169, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(209.1870, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.8118, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(76.7363, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(68.0860, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.1086, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(69.5067, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(144.0630, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(65.3335, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(67.6484, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.3488, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.1472, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(65.1886, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.2968, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.4137, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(69.7456, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(81.7870, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(137.6126, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(192.0429, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(128.6168, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(127.4624, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(105.4077, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(140.4545, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(184.6205, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(190.3819, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(199.7023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(106.6662, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(170.1346, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(180.3923, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.3879, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(61.7822, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(120.2010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(61.1927, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(70.2288, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(99.9395, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(124.2281, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.8459, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(94.7226, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.3780, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(74.9146, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(62.5617, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5576923076923077, 'recall': 0.36363636363636365, 'f1': 0.44022770398481975, 'number': 319}, 'P': {'precision': 0.5, 'recall': 0.7577937649880095, 'f1': 0.6024785510009533, 'number': 417}, 'overall_precision': 0.5142857142857142, 'overall_recall': 0.5869565217391305, 'overall_f1': 0.5482233502538071, 'overall_accuracy': 0.7198254432966912}
			------------EPOCH 14---------------
Loss:  tensor(17.0370, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(76.4783, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(117.7254, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.1008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(87.8324, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(88.7329, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(81.9626, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.8885, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(89.2995, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(101.8372, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(77.9785, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.4506, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(151.6234, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.6693, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.6584, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.9227, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1009.2314, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(415.5978, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(295.1873, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(233.8712, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(84.2556, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(115.5749, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(137.2937, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(237.3944, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(61.1013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.7027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.2784, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.2805, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.1236, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.2502, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.4303, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(75.4991, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(259.1194, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(363.1850, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(183.9451, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(112.2478, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(261.0883, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(219.7912, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(317.0601, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(260.1753, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(100.6200, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(118.3042, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(198.9162, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(326.7204, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.6439, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(76.7151, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(119.9356, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(99.3823, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(74.1525, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(94.7923, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(102.6653, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.0778, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(109.3556, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.1750, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(81.7087, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(71.0897, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5123152709359606, 'recall': 0.32601880877742945, 'f1': 0.3984674329501916, 'number': 319}, 'P': {'precision': 0.4976452119309262, 'recall': 0.7601918465227818, 'f1': 0.6015180265654649, 'number': 417}, 'overall_precision': 0.5011904761904762, 'overall_recall': 0.5720108695652174, 'overall_f1': 0.534263959390863, 'overall_accuracy': 0.7228636137656742}
			------------EPOCH 15---------------
Loss:  tensor(15.8947, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(68.1579, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(118.7508, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.8900, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(95.3559, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(95.5851, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(112.0908, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(69.9915, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(99.3194, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(231.2332, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(115.1911, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.2039, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(96.4363, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.0829, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.0502, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.6896, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(222.0741, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(61.0641, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.9294, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(70.7760, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(261.6190, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(228.1321, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(230.9321, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(145.8790, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(286.9467, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(273.2511, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(274.1295, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(524.7866, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(61.5524, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(119.6156, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(125.6903, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(170.8024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(67.8202, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(174.0588, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(146.6249, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(91.7265, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.3294, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.7916, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(103.7765, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.8634, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.3844, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.6308, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.8876, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(89.4902, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.2761, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(144.3086, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(145.0717, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(107.7400, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(199.2334, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(553.5970, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(786.5366, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(436.1672, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(779.8304, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(415.8360, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1072.0366, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1509.5774, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4325068870523416, 'recall': 0.49216300940438873, 'f1': 0.46041055718475066, 'number': 319}, 'P': {'precision': 0.5442764578833693, 'recall': 0.60431654676259, 'f1': 0.5727272727272728, 'number': 417}, 'overall_precision': 0.49515738498789347, 'overall_recall': 0.5557065217391305, 'overall_f1': 0.5236875800256082, 'overall_accuracy': 0.7065679721593107}
			------------EPOCH 16---------------
Loss:  tensor(18.3985, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(144.4399, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(91.5364, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(100.9013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.2536, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.6554, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.4347, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.1599, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(92.1104, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(102.4254, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.1668, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.8390, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(138.7779, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(62.3488, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(69.8333, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.7923, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(706.9349, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(653.1912, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(410.7111, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(99.4510, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(279.8666, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(177.4138, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(225.8731, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(393.2905, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(170.0130, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(117.0571, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(123.3736, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(110.6679, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(196.1904, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(221.7563, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(250.1359, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(436.2308, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(291.3492, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(314.3073, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(374.6032, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(286.6532, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(78.7133, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.3431, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(129.7466, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(124.1154, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(74.4156, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(99.4219, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(71.7539, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(78.9883, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.6038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.7149, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(90.3462, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.7246, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(78.5055, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(111.1437, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(110.9294, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.8421, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(200.3383, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(82.1820, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(181.3772, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(129.2595, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3611111111111111, 'recall': 0.774294670846395, 'f1': 0.49252243270189433, 'number': 319}, 'P': {'precision': 0.6535433070866141, 'recall': 0.19904076738609114, 'f1': 0.30514705882352944, 'number': 417}, 'overall_precision': 0.406905055487053, 'overall_recall': 0.4483695652173913, 'overall_f1': 0.4266321913380736, 'overall_accuracy': 0.6042644865491907}
			------------EPOCH 17---------------
Loss:  tensor(247.8307, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(554.2461, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(421.9100, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(636.6974, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(468.0905, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(390.5608, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(428.2822, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1420.2393, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(960.2169, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1007.5131, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(433.0842, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(169.6622, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(136.2422, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(107.8912, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(85.9019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(86.1087, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(363.0649, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(212.4628, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(131.1202, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(68.4499, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.7003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(95.9761, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(122.6871, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(103.1282, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(171.3323, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(146.6990, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(162.7315, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(129.9524, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(158.3951, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(215.1904, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(198.4573, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(322.0496, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(549.8295, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(408.0699, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(402.0466, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(506.4019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(390.6642, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(624.0724, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(751.6207, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(438.0731, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(228.2467, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(253.0210, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(316.1557, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(389.1418, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(77.7797, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(114.6748, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(258.6619, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(78.5900, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(108.2247, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(172.2776, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(142.6512, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(69.2377, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(189.6850, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.7361, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(84.6762, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(78.7058, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.40208333333333335, 'recall': 0.6050156739811913, 'f1': 0.48310387984981223, 'number': 319}, 'P': {'precision': 0.6197183098591549, 'recall': 0.5275779376498801, 'f1': 0.5699481865284975, 'number': 417}, 'overall_precision': 0.4946107784431138, 'overall_recall': 0.561141304347826, 'overall_f1': 0.5257797581158498, 'overall_accuracy': 0.7031431254488206}
			------------EPOCH 18---------------
Loss:  tensor(25.3480, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(139.9820, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(161.4269, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(75.0766, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(133.5713, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.9549, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(138.1367, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(88.3032, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(177.7225, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(124.6879, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(134.8744, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(114.0588, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(261.0028, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(174.3113, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(77.6269, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(90.7817, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(642.3267, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(589.4702, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(319.1242, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(99.8581, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(76.2182, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(73.3158, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(380.3046, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(317.1249, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(131.8769, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(82.3173, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(226.6393, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(168.6800, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.9863, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(87.0413, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(114.0220, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(148.3048, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(114.7404, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(155.1725, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(108.1645, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(145.8620, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(69.9346, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(83.8153, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(166.4318, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(91.6308, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.0784, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(85.1593, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(100.3820, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(120.2369, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(64.2677, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(127.2032, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(189.0531, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(86.2484, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(142.6241, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(238.0357, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(210.5657, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.8153, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(331.5966, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(71.6758, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(106.0852, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(66.3926, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.34104046242774566, 'recall': 0.36990595611285265, 'f1': 0.35488721804511275, 'number': 319}, 'P': {'precision': 0.49918166939443537, 'recall': 0.7314148681055156, 'f1': 0.5933852140077821, 'number': 417}, 'overall_precision': 0.44200626959247646, 'overall_recall': 0.5747282608695652, 'overall_f1': 0.49970466627288834, 'overall_accuracy': 0.718610175109098}
			------------EPOCH 19---------------
Loss:  tensor(17.2136, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(122.3117, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(182.0302, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(61.9779, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.2797, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.7307, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(64.2986, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(66.2488, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(68.4303, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(70.2825, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.9622, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.0379, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(66.9033, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(72.1029, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.2612, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.7085, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(159.1131, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.3548, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.6455, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.2395, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.2847, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.4262, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(87.3572, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.3947, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.6696, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.1354, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.3919, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(62.7909, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.4477, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.7253, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(70.6415, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.7545, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(68.9606, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(98.8980, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(71.8360, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.0292, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.0039, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.9325, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(85.8891, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(67.3411, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.1196, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.2594, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.0242, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(61.0968, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.3696, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(59.4540, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.9405, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.2839, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.1904, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(76.2822, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(98.0286, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.7423, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(62.6051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.2385, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.7568, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.8484, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.45866666666666667, 'recall': 0.5391849529780565, 'f1': 0.4956772334293948, 'number': 319}, 'P': {'precision': 0.5865800865800865, 'recall': 0.6498800959232613, 'f1': 0.6166097838452786, 'number': 417}, 'overall_precision': 0.5292712066905615, 'overall_recall': 0.6019021739130435, 'overall_f1': 0.5632549268912904, 'overall_accuracy': 0.7102690161851627}
			------------EPOCH 20---------------
Loss:  tensor(13.3282, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(70.9302, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(62.1238, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.4552, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.2741, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.1192, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(66.6033, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(95.6074, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(83.6092, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(75.2553, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.8397, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.5293, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.7019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(85.5012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.3176, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.1048, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(112.7884, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.0005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.5190, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.6280, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.6526, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.9896, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.1655, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.4660, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.0695, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.7519, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.6793, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.6225, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.6897, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.8880, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.8499, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.2247, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.2249, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.1288, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.6904, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.7849, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.9730, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.0303, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.1891, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.9559, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.3831, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.1414, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.2475, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.4088, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.9275, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.3692, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.5932, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.7918, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.3717, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.3651, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(68.3414, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.0389, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.8725, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.0691, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.7318, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.1725, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.42388059701492536, 'recall': 0.445141065830721, 'f1': 0.4342507645259938, 'number': 319}, 'P': {'precision': 0.5725971370143149, 'recall': 0.6714628297362111, 'f1': 0.6181015452538632, 'number': 417}, 'overall_precision': 0.5121359223300971, 'overall_recall': 0.5733695652173914, 'overall_f1': 0.541025641025641, 'overall_accuracy': 0.7289399547036403}
			------------EPOCH 21---------------
Loss:  tensor(5.4208, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.3139, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.3370, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.5459, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.3138, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.2178, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.5435, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.2705, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.0435, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.6599, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.7162, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.1047, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.5456, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.2333, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.3776, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.1574, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(81.6187, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.8350, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.6292, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.7810, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.5443, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.1688, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.1667, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.8229, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.3433, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.5550, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.5516, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.7079, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.1501, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.1067, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.0993, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.2397, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.0117, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.0609, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.5400, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.6797, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.5751, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.2514, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.2232, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.0266, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.2817, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.8330, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.4925, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.6239, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.5511, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.4023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.9371, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.6370, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.3815, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.7874, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.2336, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.0108, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.4331, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.3579, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.5902, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.6998, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4820359281437126, 'recall': 0.5047021943573667, 'f1': 0.4931087289433384, 'number': 319}, 'P': {'precision': 0.5778688524590164, 'recall': 0.6762589928057554, 'f1': 0.6232044198895027, 'number': 417}, 'overall_precision': 0.5389294403892944, 'overall_recall': 0.6019021739130435, 'overall_f1': 0.5686777920410783, 'overall_accuracy': 0.7238026846379053}
			------------EPOCH 22---------------
Loss:  tensor(4.5921, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.5835, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.5265, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.0529, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.7705, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.7310, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.2338, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.1712, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.4416, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.8865, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.8722, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.0057, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.7030, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.3280, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.9634, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.4067, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(85.3490, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.9201, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.0280, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.3345, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.8612, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.9226, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.1206, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.6146, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.4817, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.0884, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.1799, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.1855, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.5536, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.5745, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.4906, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.9963, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.2584, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.3450, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.4962, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.0934, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.6404, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.6064, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.9728, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.0056, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.7936, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.3469, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.8002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.5514, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.0984, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.4558, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.8621, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.5547, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.6953, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.1644, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.5426, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.9712, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.3585, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.5076, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.7400, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.6879, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.46407185628742514, 'recall': 0.48589341692789967, 'f1': 0.4747320061255743, 'number': 319}, 'P': {'precision': 0.5755693581780539, 'recall': 0.6666666666666666, 'f1': 0.6177777777777778, 'number': 417}, 'overall_precision': 0.5299877600979193, 'overall_recall': 0.5883152173913043, 'overall_f1': 0.557630392788152, 'overall_accuracy': 0.7296028282605093}
			------------EPOCH 23---------------
Loss:  tensor(4.0906, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.2534, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.9308, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.5525, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.6688, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.0827, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.4623, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.5041, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.2705, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.4879, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.4684, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.4300, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.7710, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.0922, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.6151, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.2871, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.8748, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.4523, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.7631, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.6791, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.9231, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.1628, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.4766, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.7861, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.8286, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.3528, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.9163, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.2014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.5469, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.8932, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.9199, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.6564, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.1056, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.6884, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.8112, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.0947, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.2358, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.3319, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.8548, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.5641, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.7195, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.8084, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.6309, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.9940, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.8347, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.4961, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.9902, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.1814, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.5054, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.1172, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.4872, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.2203, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.6688, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.5282, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.3657, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.5113, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.48059701492537316, 'recall': 0.5047021943573667, 'f1': 0.4923547400611621, 'number': 319}, 'P': {'precision': 0.5795918367346938, 'recall': 0.6810551558752997, 'f1': 0.6262403528114663, 'number': 417}, 'overall_precision': 0.5393939393939394, 'overall_recall': 0.6046195652173914, 'overall_f1': 0.5701473414477899, 'overall_accuracy': 0.7287742363144231}
			------------EPOCH 24---------------
Loss:  tensor(5.5607, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.5689, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.4567, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.4506, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.0037, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.8894, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.4626, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.8412, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.3995, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.8238, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.7894, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.3906, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.9859, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.6328, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.4884, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.8184, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.0980, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.7583, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.8651, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.5225, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.7170, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.6691, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.3173, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.4323, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.6829, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.2729, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.1410, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.9076, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.4259, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.8329, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.5894, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.3600, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.8960, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.4897, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.4039, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.7689, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.6072, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.4076, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.8992, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.8322, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.7418, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.3209, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.4945, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.1536, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.5317, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.7454, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.7183, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.0915, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.9325, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.0890, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.1745, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.3018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.6494, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.9321, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.5342, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.5265, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.47774480712166173, 'recall': 0.5047021943573667, 'f1': 0.4908536585365853, 'number': 319}, 'P': {'precision': 0.5862785862785863, 'recall': 0.6762589928057554, 'f1': 0.6280623608017817, 'number': 417}, 'overall_precision': 0.5415647921760391, 'overall_recall': 0.6019021739130435, 'overall_f1': 0.5701415701415702, 'overall_accuracy': 0.7252389106777882}
			------------EPOCH 25---------------
Loss:  tensor(5.5778, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.8165, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.5561, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.4451, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.0328, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.1968, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.9053, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.1818, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.5830, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.5299, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.9743, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.3368, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.4389, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.8703, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.8755, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.0217, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.8612, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.8048, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.3926, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.3931, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.4788, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.0802, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.2194, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.6606, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.8367, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.8726, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.0332, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.9356, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.9353, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.8531, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.4376, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.9869, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.4120, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.4280, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.4048, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.9608, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.9715, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.6992, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.3099, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.7561, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.5318, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.5422, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.9919, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.8289, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.8929, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.6833, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.8336, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.2890, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.4475, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.1691, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.2698, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.7920, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.7842, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.8757, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.4662, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.2373, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.47928994082840237, 'recall': 0.5078369905956113, 'f1': 0.4931506849315068, 'number': 319}, 'P': {'precision': 0.5805785123966942, 'recall': 0.6738609112709832, 'f1': 0.6237513873473918, 'number': 417}, 'overall_precision': 0.5389294403892944, 'overall_recall': 0.6019021739130435, 'overall_f1': 0.5686777920410783, 'overall_accuracy': 0.7239131635640501}
			------------EPOCH 26---------------
Loss:  tensor(4.4449, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.9904, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.7619, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.3878, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.8097, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.3368, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.7232, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.8574, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.6713, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.4940, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.9550, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.7788, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.8583, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.9741, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.9058, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.2876, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.0344, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.2431, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.2743, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.8709, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.3040, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.1774, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.7884, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.6366, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.7793, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.8283, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.7613, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.0323, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.4053, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.2039, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.0941, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.6145, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.4819, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.6453, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.5315, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.6640, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.6209, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.0204, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.9947, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.7150, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.7496, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.1738, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.0236, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.5628, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.0251, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.9373, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.6264, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.4605, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.2320, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.5165, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.9535, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.7014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.9602, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.4468, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.1251, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.4730, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4807121661721068, 'recall': 0.5078369905956113, 'f1': 0.4939024390243903, 'number': 319}, 'P': {'precision': 0.5805785123966942, 'recall': 0.6738609112709832, 'f1': 0.6237513873473918, 'number': 417}, 'overall_precision': 0.5395858708891595, 'overall_recall': 0.6019021739130435, 'overall_f1': 0.5690430314707771, 'overall_accuracy': 0.7229188532287466}
			------------EPOCH 27---------------
Loss:  tensor(4.1721, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.3389, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.9062, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.7059, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.9415, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.5925, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.8286, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.0552, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.6516, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.0032, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.3369, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.3621, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.1923, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.9644, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.5171, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.1478, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.2517, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.6787, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.4853, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.2955, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.4382, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.2729, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.2071, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.9931, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.9921, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.2209, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.1260, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.9117, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.9616, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.6893, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.4101, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.9086, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.7050, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.4117, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.0610, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.5504, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.5292, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.5302, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.9249, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.6611, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.4061, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.7504, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.8127, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.6666, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.4205, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.3731, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.4776, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.9361, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.3563, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.2914, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.5905, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.5310, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.2189, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.1966, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.4091, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.4583, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4725609756097561, 'recall': 0.48589341692789967, 'f1': 0.47913446676970634, 'number': 319}, 'P': {'precision': 0.5896907216494846, 'recall': 0.6858513189448441, 'f1': 0.6341463414634146, 'number': 417}, 'overall_precision': 0.5424354243542435, 'overall_recall': 0.5991847826086957, 'overall_f1': 0.5693996126533246, 'overall_accuracy': 0.724023642490195}
			------------EPOCH 28---------------
Loss:  tensor(2.3029, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.0219, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.1041, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.3952, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.1191, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.3919, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.7542, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.0641, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.4819, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.0524, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.6685, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2316, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.0684, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.4642, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.3022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.1900, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.4368, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.6214, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.7007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.1947, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6902, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.2793, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.1438, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.0285, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.0802, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.5759, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.5899, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.5155, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.4689, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.8489, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.3875, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.5248, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.3848, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.4603, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.1163, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.2430, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.2358, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.7492, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.2524, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.9993, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.0843, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.9758, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.1056, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.1679, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.4154, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.1024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.4615, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.0619, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.5166, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.6965, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.1077, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.6191, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.0281, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.5441, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.4061, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.8398, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.463855421686747, 'recall': 0.4827586206896552, 'f1': 0.47311827956989244, 'number': 319}, 'P': {'precision': 0.5878661087866108, 'recall': 0.6738609112709832, 'f1': 0.6279329608938548, 'number': 417}, 'overall_precision': 0.5370370370370371, 'overall_recall': 0.5910326086956522, 'overall_f1': 0.5627425614489004, 'overall_accuracy': 0.72573606584544}
			------------EPOCH 29---------------
Loss:  tensor(1.1717, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.3562, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.6331, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.3853, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.7513, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.0596, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.3982, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.2917, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.8106, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.6943, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.5118, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.8566, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.3440, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.2655, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.8209, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2612, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.7793, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.1209, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.7761, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.8879, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3345, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.6207, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.6354, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.3032, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.9272, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.7370, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2916, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.1112, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6923, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.0723, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.8843, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.0227, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.1456, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.8382, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.1198, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.8020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.0644, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.2973, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.4107, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.8292, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.6023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.5200, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.2986, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.5419, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.1663, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.6633, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.5294, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.7585, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.8777, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.9419, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.6692, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.7791, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.9882, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.6055, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.1397, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.3692, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4608433734939759, 'recall': 0.47962382445141066, 'f1': 0.4700460829493088, 'number': 319}, 'P': {'precision': 0.592901878914405, 'recall': 0.6810551558752997, 'f1': 0.6339285714285715, 'number': 417}, 'overall_precision': 0.5388409371146733, 'overall_recall': 0.59375, 'overall_f1': 0.5649644473173885, 'overall_accuracy': 0.7263436999392366}
			------------EPOCH 30---------------
Loss:  tensor(1.0680, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.0198, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.1001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.3577, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.3298, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.4493, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.8896, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.1060, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.8871, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.2193, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.1898, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6158, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.0769, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.6854, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.6579, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.8302, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.1333, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.6184, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.1821, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.6176, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2040, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.1217, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.2066, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.6690, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.0112, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.9695, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.8513, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.3238, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3920, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.8451, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.3070, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.3761, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.1426, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.8095, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.8193, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.3934, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.2928, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.8673, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.7892, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.0433, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.2929, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.6231, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.3739, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.6123, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.9381, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.2102, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.0497, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.7397, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.4382, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.9644, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.3186, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.2946, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.1217, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.9173, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.9812, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.8997, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.46987951807228917, 'recall': 0.4890282131661442, 'f1': 0.47926267281105994, 'number': 319}, 'P': {'precision': 0.5892116182572614, 'recall': 0.6810551558752997, 'f1': 0.631813125695217, 'number': 417}, 'overall_precision': 0.5405405405405406, 'overall_recall': 0.5978260869565217, 'overall_f1': 0.567741935483871, 'overall_accuracy': 0.7271170524222504}


		-------------RUN 4-----------
			------------EPOCH 1---------------
Loss:  tensor(1667.5442, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1072.1096, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2474.6147, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1465.9517, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1716.3810, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1266.9585, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2301.7959, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1103.4067, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1578.3604, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1577.9712, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1907.7201, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(899.0219, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1115.8870, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(748.9932, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2183.9424, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1269.8047, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1889.0652, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2681.2214, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2465.0288, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1840.4463, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1779.2502, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1869.2910, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2532.1355, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1613.2993, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1173.3457, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1553.6904, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2470.0361, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1067.4778, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(753.2467, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1301.3545, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1263.2875, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1423.2747, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1371.4546, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1145.2744, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1860.3022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2019.3628, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1728.1608, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1290.6997, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2279.7046, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1935.2777, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1203.7173, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1424.0020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(921.3197, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(790.0082, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(540.8999, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1924.5883, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(868.6342, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1805.4988, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1065.1567, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(860.5522, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1193.8094, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1598.2208, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1438.9634, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1987.2178, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1748.5715, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1157.4187, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2150.8999, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2217.4329, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.2, 'recall': 0.2602739726027397, 'f1': 0.2261904761904762, 'number': 292}, 'P': {'precision': 0.36134453781512604, 'recall': 0.25748502994011974, 'f1': 0.3006993006993007, 'number': 334}, 'overall_precision': 0.2621359223300971, 'overall_recall': 0.25878594249201275, 'overall_f1': 0.26045016077170413, 'overall_accuracy': 0.5184180382227045}
			------------EPOCH 2---------------
Loss:  tensor(1033.8799, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(719.6301, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1617.3203, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(987.0008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1031.6924, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(742.0906, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1327.4221, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(836.6741, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1363.4019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1267.9468, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1502.9238, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(706.7882, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(922.5345, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(638.0725, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1604.6458, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(875.1904, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1287.7162, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1930.4209, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1955.3831, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1446.4202, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1448.0272, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1343.3899, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1772.5515, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1206.8234, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(886.9923, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1162.5088, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1989.8098, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(834.3768, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(581.6544, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(921.5979, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(961.1610, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1071.7732, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1087.7671, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(917.5433, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1279.7325, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1567.2899, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1154.0532, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(922.7544, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1752.7988, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1366.4019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(866.1459, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1177.7506, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(690.6160, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(555.8762, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(412.5098, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1765.3154, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(615.9160, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1474.5656, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(886.2091, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(681.0264, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(977.3743, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1340.0103, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1066.9319, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1443.0420, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1341.0436, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(912.4477, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1676.4108, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1858.0887, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.24316939890710382, 'recall': 0.3047945205479452, 'f1': 0.270516717325228, 'number': 292}, 'P': {'precision': 0.32701421800947866, 'recall': 0.41317365269461076, 'f1': 0.36507936507936506, 'number': 334}, 'overall_precision': 0.2880710659898477, 'overall_recall': 0.36261980830670926, 'overall_f1': 0.3210749646393211, 'overall_accuracy': 0.5866039669620787}
			------------EPOCH 3---------------
Loss:  tensor(753.1350, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(569.5062, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1222.4838, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(803.8431, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(751.9152, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(542.4636, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(974.8358, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(655.0069, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1019.2852, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(930.2001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1184.4205, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(548.0880, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(789.8316, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(553.5042, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1334.4891, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(643.7600, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(929.8332, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1460.3125, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1534.0789, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1228.8790, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1152.6064, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(993.4317, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1355.1836, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(897.9612, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(734.7708, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(843.8834, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1435.9192, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(684.6543, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(452.2632, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(695.7061, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(810.7744, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(994.1658, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(879.5508, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(647.3972, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(900.8217, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1093.4254, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(920.6849, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(738.1016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1525.9409, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1092.7673, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(718.2122, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(965.0372, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(560.5175, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(474.3934, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(309.9350, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1484.0962, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(465.2806, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1320.3467, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(685.3179, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(473.6112, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(796.8747, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(959.3956, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(820.5071, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1207.4443, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(982.5559, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(733.3053, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1260.5702, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1209.1462, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3791208791208791, 'recall': 0.4726027397260274, 'f1': 0.42073170731707316, 'number': 292}, 'P': {'precision': 0.5067567567567568, 'recall': 0.4491017964071856, 'f1': 0.4761904761904762, 'number': 334}, 'overall_precision': 0.43636363636363634, 'overall_recall': 0.46006389776357826, 'overall_f1': 0.447900466562986, 'overall_accuracy': 0.6324229818532586}
			------------EPOCH 4---------------
Loss:  tensor(566.2333, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(377.4235, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(941.1639, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(619.8911, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(577.6413, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(508.1026, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(913.0065, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(378.7659, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(466.2915, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(417.7407, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(675.3782, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(400.0233, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(509.7313, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(408.4251, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(885.2457, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(499.1371, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(740.0260, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1210.3335, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1099.6654, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(975.6657, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(812.5380, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(671.6138, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(959.6156, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(707.3423, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(612.1971, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(611.5480, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(904.1364, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(553.6010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(324.9099, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(477.4525, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(590.1682, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(821.9468, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(641.8964, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(437.1277, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(747.4368, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(672.4346, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(726.6016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(516.2357, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1281.9174, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(844.9296, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(557.2712, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(841.5237, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(499.1101, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(327.4815, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(280.6138, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1128.7627, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(335.2000, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1251.2710, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(519.2736, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(333.1120, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(708.9517, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(640.8088, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(624.7745, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(890.6672, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(765.0028, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(555.0245, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(985.6406, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(918.3951, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.43256997455470736, 'recall': 0.5821917808219178, 'f1': 0.4963503649635036, 'number': 292}, 'P': {'precision': 0.56, 'recall': 0.41916167664670656, 'f1': 0.4794520547945206, 'number': 334}, 'overall_precision': 0.4821150855365474, 'overall_recall': 0.4952076677316294, 'overall_f1': 0.4885736800630418, 'overall_accuracy': 0.6415867848314946}
			------------EPOCH 5---------------
Loss:  tensor(434.0724, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(248.3659, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(802.5726, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(420.9162, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(346.3047, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(284.8472, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(488.1304, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(257.4539, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(331.2629, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(295.6413, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(432.6364, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(255.2258, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(309.7258, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(263.4229, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(586.7518, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(399.0331, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(575.4254, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(914.9036, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(822.5825, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(687.4159, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(524.4830, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(443.1785, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(706.6919, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(589.5970, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(441.3425, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(523.8763, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(746.5952, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(447.3817, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(306.4727, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(394.3279, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(428.9041, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(575.8013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(471.8411, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(280.8386, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(469.2399, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(424.2136, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(595.5253, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(418.1447, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(731.2214, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(823.9689, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(446.9438, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(455.2039, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(328.9160, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(336.0246, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(241.0428, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(799.0979, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(282.9223, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1008.3546, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(420.0363, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(288.2094, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(789.4756, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(577.9121, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(597.4164, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(805.8163, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(963.0661, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(652.8419, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1024.3373, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(916.6533, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.43909348441926344, 'recall': 0.5308219178082192, 'f1': 0.48062015503875966, 'number': 292}, 'P': {'precision': 0.5038363171355499, 'recall': 0.5898203592814372, 'f1': 0.543448275862069, 'number': 334}, 'overall_precision': 0.4731182795698925, 'overall_recall': 0.5623003194888179, 'overall_f1': 0.513868613138686, 'overall_accuracy': 0.6619039006450835}
			------------EPOCH 6---------------
Loss:  tensor(305.4497, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(253.3017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(645.4949, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(453.5023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(249.4658, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(207.7758, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(337.7112, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(205.8270, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(313.5105, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(243.7751, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(327.7419, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(255.1192, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(284.0957, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(329.3420, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(967.6232, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(449.9146, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(556.1978, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(965.2360, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(936.4910, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(829.8242, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(598.1510, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(484.7041, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(791.6694, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(574.6415, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(350.1884, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(388.3880, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(558.8932, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(300.1372, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(259.0951, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(345.1323, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(395.1354, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(441.1583, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(518.8411, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(374.9903, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(450.6562, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(673.7239, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(571.0543, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(415.0293, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(967.1147, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(647.8431, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(388.0111, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(479.0679, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(292.5493, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(231.4389, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(194.6874, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(658.4468, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(193.2759, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(804.6405, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(301.7331, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(219.5417, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(520.7140, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(443.3401, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(368.1774, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(523.4249, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(493.2318, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(339.3159, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(649.8142, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(539.5620, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4868913857677903, 'recall': 0.4452054794520548, 'f1': 0.46511627906976744, 'number': 292}, 'P': {'precision': 0.5046296296296297, 'recall': 0.6526946107784432, 'f1': 0.5691906005221933, 'number': 334}, 'overall_precision': 0.4978540772532189, 'overall_recall': 0.5559105431309904, 'overall_f1': 0.5252830188679246, 'overall_accuracy': 0.6875866642551396}
			------------EPOCH 7---------------
Loss:  tensor(217.1244, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(148.8088, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(450.7654, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(268.6675, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(193.6511, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(168.3742, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(280.2260, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(169.3661, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(218.7135, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(178.2078, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(236.9422, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(142.3204, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(138.9321, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(187.5774, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(403.9113, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(234.3104, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(281.0150, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(578.1940, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(482.9453, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(473.8134, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(348.2455, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(255.3992, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(422.7872, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(447.8114, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(331.3292, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(451.6661, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(415.5121, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(297.2418, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(272.0220, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(252.4229, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(293.4638, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(429.8674, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(279.4359, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(143.8097, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(172.7208, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(210.8440, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(472.7793, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(213.2697, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(372.3131, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(409.3989, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(254.1031, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(206.9593, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(286.2037, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(131.7806, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(169.7234, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(543.5461, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(117.3577, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(750.1396, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(244.8613, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(190.3178, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(600.4165, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(420.0067, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(336.7979, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(379.0375, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(448.6462, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(258.5097, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(446.2760, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(305.9828, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4978540772532189, 'recall': 0.3972602739726027, 'f1': 0.4419047619047619, 'number': 292}, 'P': {'precision': 0.4886128364389234, 'recall': 0.7065868263473054, 'f1': 0.5777233782129744, 'number': 334}, 'overall_precision': 0.49162011173184356, 'overall_recall': 0.5623003194888179, 'overall_f1': 0.5245901639344261, 'overall_accuracy': 0.67818170856695}
			------------EPOCH 8---------------
Loss:  tensor(137.1928, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(101.4983, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(318.6409, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(220.4978, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(120.9712, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(112.5652, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(171.2227, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(116.6316, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(169.1000, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(124.7698, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(160.8256, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(116.4461, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(86.0143, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(181.6740, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(321.7769, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(207.8850, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(186.6790, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(506.0129, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(389.1519, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(374.3587, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(332.7571, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(196.0396, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(296.7832, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(352.4646, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(248.5226, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(330.1762, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(279.4724, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(263.3893, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(262.0580, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(187.8762, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(226.5433, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(226.8514, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(271.9756, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(163.9771, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(122.6594, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(189.8061, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(367.8405, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(163.1985, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(303.8634, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(311.4119, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(150.3067, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(113.8523, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(169.0676, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(110.6242, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(118.7467, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(373.1668, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(98.7303, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(535.0404, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(198.3618, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(118.6505, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(454.9638, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(309.0196, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(324.5081, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(338.6559, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(414.3472, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(207.6467, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(546.1218, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(226.0866, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.6153846153846154, 'recall': 0.1643835616438356, 'f1': 0.2594594594594595, 'number': 292}, 'P': {'precision': 0.40938511326860844, 'recall': 0.7574850299401198, 'f1': 0.5315126050420168, 'number': 334}, 'overall_precision': 0.4324712643678161, 'overall_recall': 0.48083067092651754, 'overall_f1': 0.4553706505295007, 'overall_accuracy': 0.6274793512992102}
			------------EPOCH 9---------------
Loss:  tensor(163.3869, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(123.9378, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(341.0281, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(319.0586, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(132.6218, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(170.3477, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(263.4759, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(161.3295, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(129.4150, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(131.5394, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(147.9604, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(66.2020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(62.4767, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(105.3564, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(184.8133, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(125.8532, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(126.7930, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(290.5269, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(236.8918, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(249.5090, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(311.4473, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(244.9567, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(200.7084, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(295.4364, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(234.9129, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(427.0768, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(270.1892, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(250.5197, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(294.2933, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(319.1373, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(321.7717, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(415.0578, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(358.0074, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(205.0797, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(233.4577, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(266.6562, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(287.3856, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(173.0053, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(269.5221, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(372.4658, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(127.5365, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(132.7116, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(113.0902, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(136.9980, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(123.7593, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(238.7856, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(68.1899, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(391.4865, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(146.6672, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(98.1369, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(335.6911, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(211.2410, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(243.9368, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(272.2470, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(307.7470, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(170.4630, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(785.0944, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(499.5570, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.7368421052631579, 'recall': 0.0958904109589041, 'f1': 0.16969696969696968, 'number': 292}, 'P': {'precision': 0.3858858858858859, 'recall': 0.7694610778443114, 'f1': 0.514, 'number': 334}, 'overall_precision': 0.40482954545454547, 'overall_recall': 0.45527156549520764, 'overall_f1': 0.42857142857142855, 'overall_accuracy': 0.5867245433170555}
			------------EPOCH 10---------------
Loss:  tensor(364.8329, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(268.2737, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(495.8792, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(542.7622, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(228.4716, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(400.4827, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(474.3642, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(389.3971, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(419.5120, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(421.8768, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(478.3370, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(167.8519, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(175.7282, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(86.5708, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(398.6763, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(285.1793, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(210.6851, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(538.0953, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(447.9655, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(404.1018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(160.9573, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(126.6487, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(122.7958, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(223.6337, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(168.3169, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(335.9365, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(256.2529, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(246.4276, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(334.6113, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(288.8000, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(440.5928, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(437.2733, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(751.0664, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(714.5646, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(882.2792, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(957.0756, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(991.9017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(901.4126, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1306.4790, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1561.0854, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(826.2831, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(781.6842, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(343.4849, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(330.8372, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(238.1612, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(301.5832, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(150.0096, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(636.6833, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(188.7750, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(140.4520, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(196.5562, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(251.4278, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(210.1318, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(276.8319, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(231.4106, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(167.8162, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(524.1885, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(240.4589, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5869565217391305, 'recall': 0.09246575342465753, 'f1': 0.15976331360946744, 'number': 292}, 'P': {'precision': 0.37249283667621774, 'recall': 0.7784431137724551, 'f1': 0.5038759689922481, 'number': 334}, 'overall_precision': 0.385752688172043, 'overall_recall': 0.4584664536741214, 'overall_f1': 0.41897810218978104, 'overall_accuracy': 0.5820220654729608}
			------------EPOCH 11---------------
Loss:  tensor(232.1489, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(202.1986, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(435.8881, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(404.6573, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(293.3441, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(346.1438, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(539.6954, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(367.6374, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(744.5887, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(776.4203, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1151.5190, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(404.6765, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(572.8457, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(319.8494, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1040.4062, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(682., device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(804.7695, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1917.9258, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2017.6973, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1734.2583, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(778.2963, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(583.8630, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1130.5166, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(930.5330, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(561.1528, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(577.9930, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(668.8005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(452.1443, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(283.1845, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(365.7014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(430.6896, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(578.0754, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(554.7163, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(304.3350, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(598.2483, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(493.8173, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(519.8203, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(502.2106, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(685.8051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(755.2498, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(550.9171, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(642.3866, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(337.7505, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(242.2326, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(362.2319, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(655.7532, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(408.9160, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(824.3221, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(473.1788, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(365.6831, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(457.4475, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(818.3958, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1058.3368, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1290.5007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1016.0973, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(496.7523, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(962.6088, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(752.9445, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3825242718446602, 'recall': 0.6746575342465754, 'f1': 0.48822800495662955, 'number': 292}, 'P': {'precision': 0.6142131979695431, 'recall': 0.36227544910179643, 'f1': 0.455743879472693, 'number': 334}, 'overall_precision': 0.44662921348314605, 'overall_recall': 0.5079872204472844, 'overall_f1': 0.47533632286995514, 'overall_accuracy': 0.6432145656236812}
			------------EPOCH 12---------------
Loss:  tensor(280.3702, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(163.8784, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(522.1052, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(411.6039, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(286.8206, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(235.7319, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(419.1329, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(206.9330, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(280.5796, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(296.5519, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(303.6116, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(156.1989, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(201.2872, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(159.6762, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(514.0527, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(252.3511, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(356.0478, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(653.1912, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(624.9426, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(568.7939, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(390.6258, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(388.1067, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(523.7240, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(488.6561, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(366.5238, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(422.4220, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(489.0454, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(306.5405, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(231.5262, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(343.7902, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(376.8691, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(367.1070, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(410.2814, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(181.8762, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(316.9272, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(343.4162, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(464.3829, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(326.1239, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(466.3853, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(430.6689, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(288.5435, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(246.2202, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(210.1411, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(209.6231, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(144.6230, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(431.6041, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(219.0721, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(625.6230, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(259.8117, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(226.5050, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(311.5360, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(403.9895, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(363.6752, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(422.5234, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(396.7595, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(215.5926, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(327.0312, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(261.2945, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5016611295681063, 'recall': 0.5171232876712328, 'f1': 0.5092748735244519, 'number': 292}, 'P': {'precision': 0.5439024390243903, 'recall': 0.6676646706586826, 'f1': 0.5994623655913979, 'number': 334}, 'overall_precision': 0.5260196905766527, 'overall_recall': 0.597444089456869, 'overall_f1': 0.5594614809274495, 'overall_accuracy': 0.6990414179779345}
			------------EPOCH 13---------------
Loss:  tensor(114.7438, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(84.3256, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(252.5754, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(182.1367, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(136.1051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(127.3783, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(156.1801, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(117.0490, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(152.8572, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(136.1819, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(252.0468, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(196.1814, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(142.3529, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(128.2132, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(414.1052, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(195.2126, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(167.9282, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(348.4641, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(338.0091, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(478.0256, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(288.2428, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(253.2237, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(211.4817, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(428.6561, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(208.5446, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(227.0989, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(298.1782, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(131.6512, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(106.6783, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(129.5557, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(220.3116, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(160.1097, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(188.3381, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(92.8991, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(69.0210, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(148.9594, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(312.1443, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(123.1441, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(231.0426, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(249.5075, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(141.2688, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(135.5115, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(107.9766, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(94.3508, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(125.4960, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(226.8617, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(118.8014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(316.3962, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(231.4768, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(117.5143, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(211.8752, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(307.0277, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(423.7963, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(637.3278, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(355.1308, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(180.8165, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(381.5521, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(425.0628, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.501577287066246, 'recall': 0.5445205479452054, 'f1': 0.522167487684729, 'number': 292}, 'P': {'precision': 0.5872576177285319, 'recall': 0.6347305389221557, 'f1': 0.6100719424460431, 'number': 334}, 'overall_precision': 0.5471976401179941, 'overall_recall': 0.5926517571884984, 'overall_f1': 0.5690184049079754, 'overall_accuracy': 0.7085066618436124}
			------------EPOCH 14---------------
Loss:  tensor(65.6003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.9082, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(198.3758, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(106.4747, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.3207, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.0506, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(78.0218, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.9116, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(96.8823, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(66.8315, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(97.5765, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.7454, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(91.3379, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.6259, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(131.6053, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(79.9058, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(131.5997, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(249.7016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(267.4447, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(399.9238, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(250.6770, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(104.0213, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(130.2448, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(355.0327, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(307.7073, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(315.4035, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(484.6573, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(275.0914, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(185.6312, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(258.3767, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(319.2448, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(377.8391, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(256.9579, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(149.5971, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(151.7281, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(228.7844, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(276.5983, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(115.5019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(216.1040, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(188.8467, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(79.3839, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.7712, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(87.4543, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.9546, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(67.4733, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(173.3200, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.7652, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(174.8353, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(85.9338, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.4722, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(104.1002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(91.9377, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(136.4341, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(218.8454, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(139.2480, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(74.3010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(262.1343, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(216.6378, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4315068493150685, 'recall': 0.4315068493150685, 'f1': 0.4315068493150685, 'number': 292}, 'P': {'precision': 0.5441176470588235, 'recall': 0.4431137724550898, 'f1': 0.48844884488448836, 'number': 334}, 'overall_precision': 0.4858156028368794, 'overall_recall': 0.43769968051118213, 'overall_f1': 0.46050420168067224, 'overall_accuracy': 0.6572617109784771}
			------------EPOCH 15---------------
Loss:  tensor(63.9728, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(115.6255, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(423.3812, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(184.8805, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(262.5750, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(310.0364, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(521.5268, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(231.9126, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(122.8348, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(86.5673, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(102.1483, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(115.8860, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.3683, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(90.0649, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(191.7077, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(280.5855, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(131.3885, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(173.5016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(137.0105, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(143.6547, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(62.6219, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(66.7617, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(68.8533, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(100.9699, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.9109, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(118.4565, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(72.9116, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.0166, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(72.5767, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.3655, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.6298, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(77.8616, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(116.9870, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.0772, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(59.4064, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(71.4388, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(146.3927, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(71.0382, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(264.3616, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(251.9240, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(59.7824, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(367.0360, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(168.7190, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(151.4480, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(64.2903, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(365.4492, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(90.0693, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(398.0872, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(110.2740, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.7555, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(198.8046, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(115.9899, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(124.1128, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(186.6343, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(89.5754, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.3859, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(175.4835, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(123.6613, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5677966101694916, 'recall': 0.4589041095890411, 'f1': 0.5075757575757576, 'number': 292}, 'P': {'precision': 0.5252747252747253, 'recall': 0.7155688622754491, 'f1': 0.6058301647655261, 'number': 334}, 'overall_precision': 0.5397973950795948, 'overall_recall': 0.5958466453674122, 'overall_f1': 0.5664388762338649, 'overall_accuracy': 0.7019955386748659}
			------------EPOCH 16---------------
Loss:  tensor(21.5976, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.2005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(84.3959, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(88.1952, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.4764, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.7257, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.1036, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.8741, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(65.8922, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.4557, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(62.2105, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.1947, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.7942, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.6297, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.4843, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(69.0491, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(66.3787, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(137.0504, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(119.6169, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(162.9808, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(72.9884, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.1838, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.4262, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(156.3300, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(88.1431, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(149.7976, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(73.6908, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(76.7924, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.4759, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.2735, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.5038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(86.2871, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(160.2974, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.4191, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.8003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.7861, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(112.0036, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(59.0111, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(85.8693, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(125.6009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.8737, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.3445, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.1894, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.7334, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.9105, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(102.4537, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.2705, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(82.6942, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.1434, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.9148, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.8737, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.7999, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(59.5253, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(105.1593, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.8574, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.8524, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(65.2197, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.0693, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.49271137026239065, 'recall': 0.5787671232876712, 'f1': 0.5322834645669291, 'number': 292}, 'P': {'precision': 0.5885885885885885, 'recall': 0.5868263473053892, 'f1': 0.5877061469265367, 'number': 334}, 'overall_precision': 0.5399408284023669, 'overall_recall': 0.5830670926517572, 'overall_f1': 0.5606758832565285, 'overall_accuracy': 0.7062759992765418}
			------------EPOCH 17---------------
Loss:  tensor(15.6374, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.1208, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.7891, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(76.0322, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.8196, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.5950, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.9223, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.5690, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.3527, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.0531, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.7465, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.0221, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.9346, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.1808, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.3492, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.2174, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.5904, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(61.9752, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.9243, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.2886, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.3078, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.7405, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.6548, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.2320, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.5749, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(71.3672, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.2428, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.4471, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.5795, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.3950, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.3801, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.2970, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.3462, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.6525, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.2832, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.5671, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.9603, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.3794, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.1686, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(91.9400, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.9271, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.0528, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.1773, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.2228, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.9850, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.6615, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.3082, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.6205, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.1090, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.9941, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.2038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.1150, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.4173, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(76.7559, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.4505, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.1119, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.9480, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.1907, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.541958041958042, 'recall': 0.5308219178082192, 'f1': 0.5363321799307958, 'number': 292}, 'P': {'precision': 0.5635910224438903, 'recall': 0.6766467065868264, 'f1': 0.6149659863945578, 'number': 334}, 'overall_precision': 0.5545851528384279, 'overall_recall': 0.6086261980830671, 'overall_f1': 0.5803503427265803, 'overall_accuracy': 0.7164044130945921}
			------------EPOCH 18---------------
Loss:  tensor(6.4886, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.5884, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.6532, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.8797, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.3975, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.6520, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.9641, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.3680, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.7054, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.0159, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.7398, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.3683, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.3105, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.3864, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.5426, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.6692, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.5156, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.9209, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.3928, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.9312, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.8674, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.3464, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.2297, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.7290, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.5965, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.6566, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.1399, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.5314, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.9423, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.0683, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.5045, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.5049, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.1163, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.5654, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.9391, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.3162, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.7245, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.6001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.0491, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(68.0297, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.4128, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.6701, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.0142, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.1422, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.5126, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.9533, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.5092, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.3610, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.2128, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.7323, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.7250, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.6289, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.1475, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(59.6043, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.2034, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.0503, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.6162, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.5049, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5296167247386759, 'recall': 0.5205479452054794, 'f1': 0.5250431778929189, 'number': 292}, 'P': {'precision': 0.5621890547263682, 'recall': 0.6766467065868264, 'f1': 0.6141304347826086, 'number': 334}, 'overall_precision': 0.548621190130624, 'overall_recall': 0.6038338658146964, 'overall_f1': 0.5749049429657793, 'overall_accuracy': 0.7195996865014771}
			------------EPOCH 19---------------
Loss:  tensor(4.8281, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.1464, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.0584, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.3817, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.3148, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.9505, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.4567, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.1529, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.9499, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.9922, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.4947, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.9816, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.4266, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.9891, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.4379, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.0570, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.3352, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.6851, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.3096, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.4663, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.2444, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.9582, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.5061, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.2957, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.9272, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.7794, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.7932, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.7268, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.8711, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.1132, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.5109, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.5231, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.0906, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.0447, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.3095, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.2635, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.6563, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.5765, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.7529, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.2662, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.2928, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.2507, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.5227, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.9616, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.4083, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.2750, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.2486, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.3290, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.6853, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.2966, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.2039, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.3946, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.9027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.3237, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.7813, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.9506, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.2079, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.8873, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5233333333333333, 'recall': 0.5376712328767124, 'f1': 0.5304054054054055, 'number': 292}, 'P': {'precision': 0.5674300254452926, 'recall': 0.6676646706586826, 'f1': 0.6134800550206327, 'number': 334}, 'overall_precision': 0.5483405483405484, 'overall_recall': 0.6070287539936102, 'overall_f1': 0.576194086429113, 'overall_accuracy': 0.7194791101465002}
			------------EPOCH 20---------------
Loss:  tensor(3.7744, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.0918, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.1421, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.2295, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.7904, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.1663, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.5858, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.7953, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.0105, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.9730, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.8492, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.8260, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.6553, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.6526, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.8542, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.6666, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.0707, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.6496, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.6736, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.2002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.7606, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.8233, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.1247, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.7423, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.5416, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.3257, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.3865, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.1934, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.9934, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.3834, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.5465, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.8049, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.1122, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.0793, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.6430, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.8117, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.0077, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.7629, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.2799, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.7816, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.3365, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.9867, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.5090, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.6008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.6881, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.8966, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.7228, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.0488, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.9190, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.7125, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.1576, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.4682, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.2959, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.3352, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.4821, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.3977, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.4438, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.0322, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5244755244755245, 'recall': 0.5136986301369864, 'f1': 0.5190311418685121, 'number': 292}, 'P': {'precision': 0.5639097744360902, 'recall': 0.6736526946107785, 'f1': 0.6139154160982265, 'number': 334}, 'overall_precision': 0.5474452554744526, 'overall_recall': 0.5990415335463258, 'overall_f1': 0.5720823798627002, 'overall_accuracy': 0.7169470066919877}
			------------EPOCH 21---------------
Loss:  tensor(3.0758, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.2570, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.7873, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.7975, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.2500, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.7354, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.5149, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.5384, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.3543, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.4119, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.4934, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.2628, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.4873, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.6088, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.5628, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.9268, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.3264, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.1183, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.6578, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.3020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.6253, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.3243, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.1351, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.7071, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.3577, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.4748, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.5396, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.3123, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.8007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.7014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.4536, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.5831, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.6795, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.6325, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.9717, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.3909, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.8538, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.4832, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.4593, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.0600, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.4232, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.4004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.8983, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.3825, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.7255, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.9436, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.1409, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.5356, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.6931, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.1940, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.7132, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.2222, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.3871, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.6163, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.4506, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.3278, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.2886, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.4186, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5070422535211268, 'recall': 0.4931506849315068, 'f1': 0.5, 'number': 292}, 'P': {'precision': 0.5577395577395577, 'recall': 0.6796407185628742, 'f1': 0.6126855600539811, 'number': 334}, 'overall_precision': 0.5369030390738061, 'overall_recall': 0.5926517571884984, 'overall_f1': 0.5634016704631738, 'overall_accuracy': 0.7091698317959848}
			------------EPOCH 22---------------
Loss:  tensor(2.5729, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.7407, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.9674, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.3234, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.9726, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.2697, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.8308, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.9249, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.4295, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.0225, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.2897, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.8374, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.4279, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.0241, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.7359, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.5903, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.3342, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.5424, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.8340, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.5080, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.7485, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.0558, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.4832, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.2803, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.5916, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.7161, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.8724, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.8878, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.4973, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.7200, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.6508, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.7691, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.2809, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.1169, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.8410, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.6249, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.5014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.1062, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.3929, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.3887, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.2618, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.3700, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.6596, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.0733, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.9424, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.9618, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.3162, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.8796, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.5336, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5573, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.6260, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.0553, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.9046, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.8749, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.4085, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.2016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.1378, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.3051, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5172413793103449, 'recall': 0.5136986301369864, 'f1': 0.5154639175257733, 'number': 292}, 'P': {'precision': 0.5558312655086849, 'recall': 0.6706586826347305, 'f1': 0.6078697421981005, 'number': 334}, 'overall_precision': 0.5396825396825397, 'overall_recall': 0.597444089456869, 'overall_f1': 0.5670962850644428, 'overall_accuracy': 0.7125459697353349}
			------------EPOCH 23---------------
Loss:  tensor(2.2387, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.8182, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.3796, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.9678, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.6775, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.1647, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.8067, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.3085, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.3375, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2310, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.4496, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.6278, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.8999, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.8361, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.7108, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.8346, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.4751, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.9160, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.6995, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.8279, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.4825, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.5616, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.6483, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.3697, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.6216, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.0750, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.3614, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.4408, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.6492, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.2060, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.0828, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.6979, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.8725, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2382, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.4757, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.9935, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.8386, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.6725, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.1717, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.5645, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.9373, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.4010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.4439, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.9667, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.4026, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.5912, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.2221, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.6288, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.5624, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.1493, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.4059, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.8699, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.7671, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.9728, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.7416, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.5785, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.8431, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.1716, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5121107266435986, 'recall': 0.5068493150684932, 'f1': 0.5094664371772805, 'number': 292}, 'P': {'precision': 0.5533498759305211, 'recall': 0.6676646706586826, 'f1': 0.6051560379918588, 'number': 334}, 'overall_precision': 0.5361271676300579, 'overall_recall': 0.5926517571884984, 'overall_f1': 0.5629742033383915, 'overall_accuracy': 0.713028275155242}
			------------EPOCH 24---------------
Loss:  tensor(1.9122, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.0974, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.0431, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.5859, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.8858, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.5073, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.6962, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.9521, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.9374, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.8305, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.3802, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.5306, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.1788, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.3233, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.7467, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.1785, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.4043, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.9225, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.9692, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.9123, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.4028, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.0434, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.0383, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.4263, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.7906, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.4387, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.4507, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.0954, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.8411, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.5281, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.2715, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.4916, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.5141, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.7851, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.0272, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.2649, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.5993, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.3149, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.6984, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.2073, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.2990, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.7724, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.5495, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.8579, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.8589, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.1019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.2146, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.1914, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.7651, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.8079, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.5232, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.1371, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.2710, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.1949, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.7581, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.7999, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.8831, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.2264, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5103448275862069, 'recall': 0.5068493150684932, 'f1': 0.5085910652920963, 'number': 292}, 'P': {'precision': 0.551980198019802, 'recall': 0.6676646706586826, 'f1': 0.6043360433604337, 'number': 334}, 'overall_precision': 0.5345821325648416, 'overall_recall': 0.5926517571884984, 'overall_f1': 0.5621212121212121, 'overall_accuracy': 0.711641647073009}
			------------EPOCH 25---------------
Loss:  tensor(1.6548, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.7826, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.7644, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.7755, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.3416, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.0168, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.9706, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.7094, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.8368, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6081, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.6500, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.5689, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.6067, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.0020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.1587, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.6979, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.6857, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.9564, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.1189, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.4304, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.8618, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.6338, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.4571, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.3313, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.3881, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.2677, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.6041, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.8133, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5398, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.0359, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.6172, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.8625, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.9050, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5483, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.7045, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.9348, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.5850, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.9504, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.8032, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.3935, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.7526, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.4041, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.8682, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2518, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5723, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.8736, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.8842, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.2381, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.2654, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.8376, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.5310, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.3385, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.0113, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.5753, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.3976, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.3017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.7138, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5033783783783784, 'recall': 0.5102739726027398, 'f1': 0.5068027210884354, 'number': 292}, 'P': {'precision': 0.5561097256857855, 'recall': 0.6676646706586826, 'f1': 0.6068027210884355, 'number': 334}, 'overall_precision': 0.533715925394548, 'overall_recall': 0.5942492012779552, 'overall_f1': 0.562358276643991, 'overall_accuracy': 0.7097727135708688}
			------------EPOCH 26---------------
Loss:  tensor(1.4804, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5843, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.0486, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.3321, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.9024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.6357, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.4829, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5174, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.9540, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4163, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.1820, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.7415, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.1944, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.7782, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.7498, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.3373, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.1627, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.1664, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.5087, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.4149, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.4959, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.3610, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.0743, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.7953, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.7188, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.6223, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.8804, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2025, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.6607, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2461, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.9938, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.4369, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3783, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.4186, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.1152, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.8250, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.3070, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.2153, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.2550, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.3961, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.1190, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.4316, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.9170, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2669, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.0922, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6330, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.4259, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.8382, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4108, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.3402, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.1112, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.5192, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.2771, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.9736, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.0600, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.7612, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.2270, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5068027210884354, 'recall': 0.5102739726027398, 'f1': 0.5085324232081911, 'number': 292}, 'P': {'precision': 0.5506172839506173, 'recall': 0.6676646706586826, 'f1': 0.6035182679296347, 'number': 334}, 'overall_precision': 0.5321888412017167, 'overall_recall': 0.5942492012779552, 'overall_f1': 0.561509433962264, 'overall_accuracy': 0.7121842406704045}
			------------EPOCH 27---------------
Loss:  tensor(1.3267, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4095, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.5632, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.9877, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5482, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.3368, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.0698, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3477, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.2878, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2580, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.8193, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.9966, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.8491, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5973, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.4094, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.0430, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.7522, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.5305, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.0700, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.6237, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2131, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.1466, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.7929, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.3520, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.1530, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.5445, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.1939, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4344, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.9357, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.3589, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.9775, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.2962, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.9722, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2496, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.1749, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.5553, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.2275, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.1219, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.1127, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.8475, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.1138, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.8938, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.1059, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6796, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.0401, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.5200, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4530, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.8286, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5143, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2651, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.9376, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.7685, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.9386, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.0126, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.6040, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.8304, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.3597, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.8683, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5050847457627119, 'recall': 0.5102739726027398, 'f1': 0.5076660988074958, 'number': 292}, 'P': {'precision': 0.5506172839506173, 'recall': 0.6676646706586826, 'f1': 0.6035182679296347, 'number': 334}, 'overall_precision': 0.5314285714285715, 'overall_recall': 0.5942492012779552, 'overall_f1': 0.5610859728506787, 'overall_accuracy': 0.711641647073009}
			------------EPOCH 28---------------
Loss:  tensor(1.2051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2709, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.2167, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.7500, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2618, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.0922, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.7572, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2200, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.8095, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1459, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5571, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.3621, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5730, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4541, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.1600, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.8207, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.4475, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.0182, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.7454, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.9922, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.0028, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.9732, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5592, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.6451, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.4110, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.7395, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.5382, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2938, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.7161, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.0762, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.7701, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.7151, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.5975, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1452, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.9270, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.1567, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.7589, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.9846, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.2438, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.7889, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.8912, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.7215, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.8566, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4926, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.8493, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.0721, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3140, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.3608, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2497, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1424, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.6133, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.4858, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.4694, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.0959, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.3214, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6484, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.0296, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5722, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5066666666666667, 'recall': 0.5205479452054794, 'f1': 0.5135135135135136, 'number': 292}, 'P': {'precision': 0.5597014925373134, 'recall': 0.6736526946107785, 'f1': 0.6114130434782609, 'number': 334}, 'overall_precision': 0.5370370370370371, 'overall_recall': 0.6022364217252396, 'overall_f1': 0.5677710843373494, 'overall_accuracy': 0.7155603786097546}
			------------EPOCH 29---------------
Loss:  tensor(1.1032, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1561, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.9439, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5671, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.0229, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.8943, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1159, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.4510, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0555, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.3511, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.7835, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.3460, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3360, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.9671, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6445, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2090, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.5812, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.4853, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.4810, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.8317, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.8280, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.3488, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.9276, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.1945, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.1226, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.9129, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1784, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5695, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.8271, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5964, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.2174, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.3939, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0522, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.7118, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.8162, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.3652, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.6531, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.5113, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.9266, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.7123, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5626, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6566, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3325, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6982, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.6935, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1977, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.0095, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.0334, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0413, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.3473, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2601, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.1033, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.1279, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.1075, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4956, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.7632, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.3427, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5016501650165016, 'recall': 0.5205479452054794, 'f1': 0.5109243697478991, 'number': 292}, 'P': {'precision': 0.5639097744360902, 'recall': 0.6736526946107785, 'f1': 0.6139154160982265, 'number': 334}, 'overall_precision': 0.5370370370370371, 'overall_recall': 0.6022364217252396, 'overall_f1': 0.5677710843373494, 'overall_accuracy': 0.7126062579128233}
			------------EPOCH 30---------------
Loss:  tensor(1.0185, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0598, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.7054, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4138, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.8308, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.7295, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2795, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0288, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.1548, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9817, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.1522, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.2581, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.1210, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2253, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.7646, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4772, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.9769, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.1706, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2473, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.0370, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6654, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6971, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.1973, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.1682, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.7698, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.5613, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.2314, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0820, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4060, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.6309, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4669, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.8406, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.4854, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9645, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5459, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5422, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.0443, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.3809, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.9173, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.2330, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5656, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4317, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4974, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2037, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5575, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.3601, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1076, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.7066, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.8403, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9491, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.1305, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.0672, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.7688, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.7724, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.9160, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3674, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5229, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.1301, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5033557046979866, 'recall': 0.5136986301369864, 'f1': 0.5084745762711865, 'number': 292}, 'P': {'precision': 0.5555555555555556, 'recall': 0.6736526946107785, 'f1': 0.6089309878213802, 'number': 334}, 'overall_precision': 0.5334281650071123, 'overall_recall': 0.5990415335463258, 'overall_f1': 0.564334085778781, 'overall_accuracy': 0.7126665460903117}


		-------------RUN 5-----------
			------------EPOCH 1---------------
Loss:  tensor(2350.2437, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3649.6763, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1430.9990, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2585.9443, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2231.3450, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1867.8232, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2171.1711, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1722.2632, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2660.1997, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1799.0316, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1679.9497, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1432.2506, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1889.8364, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1498.0815, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1408.0229, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2317.3530, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1246.8596, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1504.8408, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2763.2451, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2174.1848, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2428.3899, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2679.7456, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3286.9634, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2546.1641, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1454.3528, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1746.7803, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2155.3271, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1844.3088, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(889.9379, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1601.7030, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1894.8252, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1719.1763, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1462.9968, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2550.5952, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1910.2432, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1551.8861, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1285.5256, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1170.0936, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1709.9470, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1241.7905, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1605.4407, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1805.3606, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(855.6514, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1754.6622, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1029.9000, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(766.0120, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(951.7036, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(966.9961, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2646.2295, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1012.5402, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1878.2911, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(619.2448, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(797.7021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(826.6444, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1074.8717, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(711.9969, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.0763888888888889, 'recall': 0.03303303303303303, 'f1': 0.046121593291404604, 'number': 333}, 'P': {'precision': 0.05333333333333334, 'recall': 0.013468013468013467, 'f1': 0.02150537634408602, 'number': 297}, 'overall_precision': 0.0684931506849315, 'overall_recall': 0.023809523809523808, 'overall_f1': 0.0353356890459364, 'overall_accuracy': 0.3827440599987395}
			------------EPOCH 2---------------
Loss:  tensor(1418.3865, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2501.2051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(893.8705, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1823.4480, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1482.1841, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1271.6837, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1466.3459, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1148.9517, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1986.7683, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1285.1124, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1086.1965, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(935.3010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1381.8628, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(984.3057, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(934.7274, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1554.3961, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(839.8470, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1209.3362, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2329.1919, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1748.4716, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1853.8179, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1911.5157, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2566.2998, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1905.4265, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(992.3431, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1487.1112, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1679.6555, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1505.6768, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(713.3796, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1205.5917, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1426.7349, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1148.3384, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1091.9205, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1956.2317, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1369.6832, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1155.5411, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(988.4696, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(957.6635, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1165.3740, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(919.8364, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1182.5747, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1503.4238, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(668.1007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1348.2510, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(813.3633, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(629.0514, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(777.5276, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(746.9523, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2326.5300, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(813.2590, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1702.3905, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(445.0562, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(632.3602, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(653.7500, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(899.0938, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(486.7817, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.0989399293286219, 'recall': 0.08408408408408409, 'f1': 0.09090909090909091, 'number': 333}, 'P': {'precision': 0.19557195571955718, 'recall': 0.17845117845117844, 'f1': 0.18661971830985916, 'number': 297}, 'overall_precision': 0.14620938628158844, 'overall_recall': 0.12857142857142856, 'overall_f1': 0.13682432432432431, 'overall_accuracy': 0.5027415390432974}
			------------EPOCH 3---------------
Loss:  tensor(1007.8337, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1878.8104, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(716.8827, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1254.9369, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1116.3119, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1030.9229, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1151.2249, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(947.1493, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1670.4766, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(947.6369, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(702.7607, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(793.2552, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1234.1406, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(910.0298, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(835.5342, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1397.2611, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(668.5236, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1002.1676, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1663.2314, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1383.8711, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1610.4204, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1480.8074, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2232.2681, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1517.2069, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(781.2915, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1331.6499, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1436.7413, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1265.5492, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(604.8092, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1045.7087, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1207.5801, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(951.7570, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(943.0009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1660.2234, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1059.9417, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(864.4468, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(779.7203, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(785.7986, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(963.8333, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(770.0242, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1037.0360, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1214.1334, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(544.0383, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1076.6437, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(637.3615, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(494.3367, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(610.8241, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(576.0785, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1951.3362, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(626.7564, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1509.9633, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(359.0597, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(530.2962, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(538.7972, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(780.1141, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(387.0717, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.18895348837209303, 'recall': 0.19519519519519518, 'f1': 0.19202363367799113, 'number': 333}, 'P': {'precision': 0.3258426966292135, 'recall': 0.19528619528619529, 'f1': 0.2442105263157895, 'number': 297}, 'overall_precision': 0.23563218390804597, 'overall_recall': 0.19523809523809524, 'overall_f1': 0.21354166666666669, 'overall_accuracy': 0.5077204260414697}
			------------EPOCH 4---------------
Loss:  tensor(890.5526, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1663.0864, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(544.6538, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1031.6113, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(961.4628, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(874.6163, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(972.3912, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(775.3698, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1327.9402, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(692.6332, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(611.6689, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(681.4530, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1039.2736, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(781.5062, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(720.5072, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1213.3306, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(533.5763, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(793.3741, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1201.0098, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1131.9072, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1357.5618, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1196.6416, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1915.1846, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1253.1624, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(643.4304, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1193.8840, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1236.9436, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1065.5905, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(537.8995, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(811.2094, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(991.6003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(746.7158, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(779.2949, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1336.7775, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(904.3969, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(685.8923, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(649.7581, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(636.2723, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(749.8032, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(608.8560, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(754.3090, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(942.8807, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(426.2404, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(852.7047, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(480.9945, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(363.9803, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(446.7529, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(443.7730, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1482.6960, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(500.4307, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1295.5410, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(273.3095, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(427.8319, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(425.1156, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(715.5948, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(322.0765, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3642691415313225, 'recall': 0.47147147147147145, 'f1': 0.4109947643979058, 'number': 333}, 'P': {'precision': 0.44776119402985076, 'recall': 0.40404040404040403, 'f1': 0.4247787610619469, 'number': 297}, 'overall_precision': 0.39628040057224606, 'overall_recall': 0.4396825396825397, 'overall_f1': 0.41685477802859294, 'overall_accuracy': 0.5857439969748535}
			------------EPOCH 5---------------
Loss:  tensor(672.9855, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1152.7454, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(371.5286, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(656.2095, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(725.9153, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(630.4577, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(791.4108, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(552.3000, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1057.6616, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(539.0134, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(494.3885, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(537.9282, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(883.6125, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(596.3987, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(562.9700, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(970.6650, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(492.3989, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(558.5864, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(805.5118, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(827.8502, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1124.2163, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(924.0040, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1668.3428, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(897.8558, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(513.0414, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(983.9250, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(992.0084, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(780.9655, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(445.8500, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(663.7341, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(806.6130, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(484.6403, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(624.6960, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1215.2452, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(774.1903, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(538.4423, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(630.1701, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(635.9055, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(606.7207, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(416.3093, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(610.3000, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(864.7112, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(315.9948, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(615.0073, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(299.9763, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(261.7690, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(292.8604, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(348.3214, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1050.8051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(426.6714, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(942.6582, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(190.9942, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(332.5288, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(413.8150, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(688.9789, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(247.4543, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.42786069651741293, 'recall': 0.25825825825825827, 'f1': 0.32209737827715357, 'number': 333}, 'P': {'precision': 0.40130151843817785, 'recall': 0.622895622895623, 'f1': 0.4881266490765172, 'number': 297}, 'overall_precision': 0.4093655589123867, 'overall_recall': 0.4301587301587302, 'overall_f1': 0.41950464396284826, 'overall_accuracy': 0.5968362009201488}
			------------EPOCH 6---------------
Loss:  tensor(561.2263, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(983.7070, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(277.0130, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(655.7114, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(662.7691, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(514.7078, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(528.3062, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(485.3223, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(751.5378, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(376.5581, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(338.6956, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(382.0687, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(738.5204, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(437.4014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(448.3128, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(810.2896, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(477.9460, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(587.3911, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(643.1451, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(583.8298, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(902.5613, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(739.4078, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1206.2427, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(623.5598, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(342.8507, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(688.0425, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(621.8253, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(466.2895, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(465.1371, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(638.8290, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(776.2817, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(367.0130, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(566.6460, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1792.4541, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1064.2756, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(656.6056, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(437.2979, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(406.8190, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(537.0284, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(420.0953, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(459.2965, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(525.7916, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(228.7488, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(406.3457, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(270.8837, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(253.7561, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(250.9441, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(441.6238, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1372.8972, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(562.9861, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1127.6215, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(261.4716, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(490.8387, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(479.3337, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(639.0867, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(217.3925, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.42789598108747046, 'recall': 0.5435435435435435, 'f1': 0.47883597883597884, 'number': 333}, 'P': {'precision': 0.5396039603960396, 'recall': 0.367003367003367, 'f1': 0.43687374749499003, 'number': 297}, 'overall_precision': 0.464, 'overall_recall': 0.4603174603174603, 'overall_f1': 0.46215139442231074, 'overall_accuracy': 0.6491460263439843}
			------------EPOCH 7---------------
Loss:  tensor(497.2626, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(923.2551, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(330.2129, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(504.7635, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(529.7947, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(349.0176, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(426.0910, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(332.6769, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(517.4804, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(314.9696, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(270.0199, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(349.1990, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(717.6927, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(584.0720, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(475.0722, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(789.6290, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(411.6714, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(429.6549, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(597.6581, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(700.1525, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(722.3073, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(551.5630, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1047.3750, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(520.6494, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(274.5211, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(571.2805, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(514.4557, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(430.2559, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(282.1336, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(481.3984, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(642.7440, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(367.8395, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(337.9771, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(723.8350, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(465.1089, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(302.0645, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(301.9623, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(297.8886, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(341.3416, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(244.8788, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(384.6160, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(434.5245, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(171.6360, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(311.9380, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(131.6131, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(126.8080, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(124.3675, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(247.5897, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(751.7493, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(207.8073, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(673.9115, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(123.2492, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(212.3027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(225.2149, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(324.2080, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(107.7586, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4792243767313019, 'recall': 0.5195195195195195, 'f1': 0.49855907780979825, 'number': 333}, 'P': {'precision': 0.5571428571428572, 'recall': 0.5252525252525253, 'f1': 0.5407279029462739, 'number': 297}, 'overall_precision': 0.5132605304212169, 'overall_recall': 0.5222222222222223, 'overall_f1': 0.5177025963808025, 'overall_accuracy': 0.6677380727295645}
			------------EPOCH 8---------------
Loss:  tensor(299.3633, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(451.8354, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(194.8735, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(301.8271, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(473.6792, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(248.7548, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(290.6019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(239.1918, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(400.2628, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(217.2641, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(169.9879, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(187.6954, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(423.0759, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(282.2666, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(266.4494, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(624.4739, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(188.4728, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(224.0790, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(316.8713, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(322.5626, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(502.8954, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(363.3827, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(709.2139, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(242.0850, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(141.5125, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(384.8017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(268.6299, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(228.0860, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(190.7188, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(285.7315, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(294.8862, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(114.3465, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(172.9015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(500.2395, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(303.8314, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(135.4267, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(227.3301, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(221.2747, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(239.0504, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(170.4837, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(323.8866, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(343.1741, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(135.1657, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(212.5516, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(97.9276, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(103.0178, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(96.1107, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(150.7632, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(494.9688, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(138.9631, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(536.7346, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(107.4195, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(153.0352, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(173.1013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(285.1843, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(94.8933, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5045662100456622, 'recall': 0.6636636636636637, 'f1': 0.5732814526588845, 'number': 333}, 'P': {'precision': 0.5201465201465202, 'recall': 0.4781144781144781, 'f1': 0.4982456140350877, 'number': 297}, 'overall_precision': 0.510548523206751, 'overall_recall': 0.5761904761904761, 'overall_f1': 0.5413870246085012, 'overall_accuracy': 0.6712043864624693}
			------------EPOCH 9---------------
Loss:  tensor(289.3291, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(404.9786, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(139.7160, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(208.5869, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(297.2188, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(157.4682, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(288.9862, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(177.7332, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(238.1986, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(137.7272, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(85.9846, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(98.6380, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(237.9853, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(184.2755, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(161.2218, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(377.9611, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(142.1147, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(143.3233, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(187.5648, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(161.6673, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(477.2224, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(376.3543, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(812.2200, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(280.5956, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(101.6119, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(394.8945, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(189.1375, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(190.8884, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(149.4357, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(199.9113, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(222.3219, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(67.3466, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(115.9355, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(360.0678, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(226.5509, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(78.5495, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(151.1682, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(135.9629, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(189.3105, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(121.2587, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(266.6496, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(286.2920, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(123.1260, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(193.8277, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(122.6745, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(91.6504, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(106.9086, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(147.7525, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(704.5643, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(130.6140, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(452.1178, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(80.4541, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(168.3259, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(122.9348, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(275.2159, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(96.9619, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4485776805251641, 'recall': 0.6156156156156156, 'f1': 0.5189873417721518, 'number': 333}, 'P': {'precision': 0.5015974440894568, 'recall': 0.5286195286195287, 'f1': 0.5147540983606557, 'number': 297}, 'overall_precision': 0.4701298701298701, 'overall_recall': 0.5746031746031746, 'overall_f1': 0.5171428571428571, 'overall_accuracy': 0.6485157874834563}
			------------EPOCH 10---------------
Loss:  tensor(255.5141, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(340.2611, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(146.1397, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(229.8769, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(211.6353, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(119.8924, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(307.5576, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(154.0703, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(248.3654, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(141.0856, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(71.4896, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(79.5354, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(162.1539, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(151.5645, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(134.8597, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(301.8771, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(113.7941, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(137.8862, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(143.8425, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(103.6652, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(324.5956, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(275.6143, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(373.1292, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(131.6949, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(185.9142, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(328.2151, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(288.1925, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(355.3036, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(249.6803, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(198.0942, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(204.4525, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(127.1194, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(388.0995, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(403.7640, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(327.5931, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(124.3846, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(206.4001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(134.2272, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(255.6799, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(209.4816, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(218.5341, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(158.7164, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(88.5784, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(161.2505, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(76.6850, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(65.5413, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(77.1254, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(87.5123, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(299.4125, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(100.7660, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(338.5972, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.3218, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(75.2990, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(97.6930, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(178.8888, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(71.3247, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3552397868561279, 'recall': 0.6006006006006006, 'f1': 0.44642857142857145, 'number': 333}, 'P': {'precision': 0.42792792792792794, 'recall': 0.31986531986531985, 'f1': 0.36608863198458574, 'number': 297}, 'overall_precision': 0.37579617834394907, 'overall_recall': 0.46825396825396826, 'overall_f1': 0.41696113074204944, 'overall_accuracy': 0.5800088233440474}
			------------EPOCH 11---------------
Loss:  tensor(463.0828, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(656.7454, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(410.3620, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(342.4586, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(311.1955, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(308.3314, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(818.7353, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(417.3768, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1611.0220, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1072.0591, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(279.2765, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(153.4044, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(249.7313, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(219.2412, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(171.7822, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(328.4667, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(99.4744, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(170.3878, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(204.6534, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(157.3992, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(274.8687, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(218.3716, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(339.1218, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(102.0075, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(100.5508, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(162.6697, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(131.0885, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(109.4150, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(126.4567, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(166.7864, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(261.9148, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(170.3631, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(337.1038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(479.4183, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(299.0902, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(174.2926, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(257.2106, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(161.1031, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(556.3475, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(340.5734, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(339.8824, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(286.8567, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(219.2620, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(476.8879, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(190.2658, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(138.8004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(159.5738, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(150.8564, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(533.0578, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(221.1001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(553.9873, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(110.3950, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(160.6643, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(280.3571, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(280.0188, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(158.7346, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.47058823529411764, 'recall': 0.6486486486486487, 'f1': 0.5454545454545454, 'number': 333}, 'P': {'precision': 0.5082508250825083, 'recall': 0.5185185185185185, 'f1': 0.5133333333333333, 'number': 297}, 'overall_precision': 0.48556430446194226, 'overall_recall': 0.5873015873015873, 'overall_f1': 0.5316091954022988, 'overall_accuracy': 0.6746076763093213}
			------------EPOCH 12---------------
Loss:  tensor(228.7606, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(285.0454, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(95.6023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(161.2857, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(167.6112, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(106.0650, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(193.3826, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(106.6444, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(364.0616, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(235.7898, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(235.4384, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(121.2508, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(410.2252, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(224.3589, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(326.9235, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(378.5022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(212.8266, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(294.0171, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(484.0761, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(389.8551, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(430.8504, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(316.2002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(400.7205, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(222.5579, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(111.4450, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(362.9813, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(253.6333, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(495.3764, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(206.2984, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(268.3313, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(536.8206, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(230.7909, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(217.6956, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(568.2483, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(329.6389, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(106.5007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(143.2672, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(124.0103, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(459.8635, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(279.7912, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(176.2385, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(115.6806, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.7652, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(95.3999, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(83.1750, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(64.4729, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(79.2013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(77.5067, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(191.5555, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(74.8394, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(188.4427, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.0484, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(84.4152, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(195.4211, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(287.8247, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(109.5583, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.47665847665847666, 'recall': 0.5825825825825826, 'f1': 0.5243243243243243, 'number': 333}, 'P': {'precision': 0.5545454545454546, 'recall': 0.4107744107744108, 'f1': 0.4719535783365571, 'number': 297}, 'overall_precision': 0.5039872408293461, 'overall_recall': 0.5015873015873016, 'overall_f1': 0.5027844073190135, 'overall_accuracy': 0.6874015251780424}
			------------EPOCH 13---------------
Loss:  tensor(224.7544, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(306.4780, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(105.7162, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(269.3263, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(175.0566, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(99.3006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(119.4185, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(98.3917, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(136.7446, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(77.1588, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(61.1572, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(67.8441, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(119.8959, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(100.1527, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(101.6824, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(174.7571, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(80.3298, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(88.9543, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(119.1175, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(84.7777, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(208.3662, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(203.3499, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(360.9399, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(128.7650, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(80.6019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(153.7190, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(134.6772, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(119.4884, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(77.6590, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(136.5780, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(131.5250, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.3841, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(68.4161, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(162.3413, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(84.4613, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.3626, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(78.1417, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(65.6380, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(103.7646, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(74.2867, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(115.9460, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(105.3704, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(69.2404, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(107.3850, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(61.4654, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(64.8712, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(68.0431, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(61.2357, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(134.5119, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(93.5030, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(195.7895, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.4825, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(61.1608, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(114.2072, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(95.8894, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(64.7688, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5804878048780487, 'recall': 0.35735735735735735, 'f1': 0.44237918215613387, 'number': 333}, 'P': {'precision': 0.4613861386138614, 'recall': 0.7845117845117845, 'f1': 0.5810473815461347, 'number': 297}, 'overall_precision': 0.49577464788732395, 'overall_recall': 0.5587301587301587, 'overall_f1': 0.5253731343283583, 'overall_accuracy': 0.6686204071343039}
			------------EPOCH 14---------------
Loss:  tensor(248.4327, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(484.3539, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(173.9988, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(282.8667, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(175.4928, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.0592, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(96.2721, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(110.7324, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(90.9677, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.1148, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.4049, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.0482, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(94.3506, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(79.8836, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(71.1548, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(142.0696, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(79.5933, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(78.4497, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(130.9826, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(118.6859, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(401.6930, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(451.1182, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(413.8622, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(182.4522, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(69.2552, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(307.7809, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(218.6161, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(189.9482, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(109.1696, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(85.4236, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(119.1576, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.5197, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.2351, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(119.5186, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(75.1157, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.4757, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.1903, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.2761, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.7589, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.5765, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(74.2977, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(64.4333, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.2232, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.4192, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.7549, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.6323, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.8855, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.5810, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(148.2866, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.8136, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(194.5349, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(128.0149, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.6502, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(94.4807, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(145.4034, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(94.8915, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.7339449541284404, 'recall': 0.24024024024024024, 'f1': 0.36199095022624433, 'number': 333}, 'P': {'precision': 0.4128113879003559, 'recall': 0.7811447811447811, 'f1': 0.540162980209546, 'number': 297}, 'overall_precision': 0.46497764530551416, 'overall_recall': 0.49523809523809526, 'overall_f1': 0.4796310530361261, 'overall_accuracy': 0.6077393332072856}
			------------EPOCH 15---------------
Loss:  tensor(229.7237, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(314.4048, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(71.3578, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(376.2589, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(263.0813, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(360.6858, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(139.1985, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(315.5258, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(103.6668, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(178.4971, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(82.9947, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.8528, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(134.1234, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(101.8938, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(112.0808, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(213.0040, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.8321, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.8187, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.1796, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.5620, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(151.1020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(163.8839, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(141.5821, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.5655, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(148.8619, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(183.3367, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(231.9452, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(169.3106, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(400.6193, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(842.8303, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1431.1711, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1682.1604, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(153.4853, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(203.3901, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(304.0983, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(214.1603, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.4348, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(126.5904, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(81.1789, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.2692, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(72.2863, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(76.2619, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.2318, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.7540, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.3929, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.4037, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.1827, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.2134, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(205.7674, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(88.1752, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(342.1497, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.0693, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(116.7202, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(184.4297, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(208.0738, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(110.8134, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.8253968253968254, 'recall': 0.15615615615615616, 'f1': 0.2626262626262626, 'number': 333}, 'P': {'precision': 0.36977491961414793, 'recall': 0.7744107744107744, 'f1': 0.500544069640914, 'number': 297}, 'overall_precision': 0.4116788321167883, 'overall_recall': 0.44761904761904764, 'overall_f1': 0.42889733840304184, 'overall_accuracy': 0.5575723199092456}
			------------EPOCH 16---------------
Loss:  tensor(288.3097, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1116.1125, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(303.0435, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(808.4137, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(622.2567, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(135.9241, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(207.2650, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(548.0016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(926.8560, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(491.2773, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(224.6876, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1031.3698, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1066.3092, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(582.5682, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(370.5746, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(616.6167, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(95.8304, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(219.0220, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(145.5456, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(289.2604, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(263.8524, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(212.4258, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(316.5135, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(118.8654, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(100.7469, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(206.6460, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(142.0581, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(165.5546, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(133.6728, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(288.8380, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(254.6039, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(286.5074, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(445.9228, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(706.5928, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(669.4849, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(445.4550, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(494.9078, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(679.5107, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(773.1044, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(540.2845, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(690.2690, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1554.3254, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(520.9822, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(957.8248, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(279.4245, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(287.7324, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(273.5917, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(345.2426, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(363.9602, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(189.5293, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(407.3983, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(218.7876, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(138.9453, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(262.6059, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(296.2547, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(115.2234, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5714285714285714, 'recall': 0.5405405405405406, 'f1': 0.5555555555555556, 'number': 333}, 'P': {'precision': 0.5223880597014925, 'recall': 0.5892255892255892, 'f1': 0.5537974683544304, 'number': 297}, 'overall_precision': 0.5461538461538461, 'overall_recall': 0.5634920634920635, 'overall_f1': 0.5546874999999999, 'overall_accuracy': 0.7249007373794668}
			------------EPOCH 17---------------
Loss:  tensor(203.3877, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(208.8967, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(91.9448, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(131.1457, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(146.7693, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(85.9644, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(118.9252, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(111.6807, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(123.0875, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(95.4153, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.5053, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(88.5972, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(405.4662, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(206.4310, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(164.6484, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(367.9583, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(260.0261, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(236.3191, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(247.0345, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(418.6390, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(706.4253, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(402.4598, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(872.6885, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(486.7400, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(349.2305, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(413.9889, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(471.4338, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(423.2214, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(122.0908, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(157.7253, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(154.1609, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(71.0335, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(84.5334, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(299.2332, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(134.3637, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(97.6106, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(74.4906, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(66.8084, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(64.6120, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.7206, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(80.6909, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(92.2792, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.2005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.8197, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.6961, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.7686, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(71.5801, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(115.6805, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(103.5717, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.2572, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(138.8206, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(96.1186, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(85.5306, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(127.3102, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(182.9077, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(76.2248, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.45521023765996343, 'recall': 0.7477477477477478, 'f1': 0.5659090909090909, 'number': 333}, 'P': {'precision': 0.5257142857142857, 'recall': 0.30976430976430974, 'f1': 0.38983050847457623, 'number': 297}, 'overall_precision': 0.47229916897506924, 'overall_recall': 0.5412698412698412, 'overall_f1': 0.5044378698224852, 'overall_accuracy': 0.6400075628663263}
			------------EPOCH 18---------------
Loss:  tensor(488.7314, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(660.5251, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(153.1092, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(295.6810, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(306.8416, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(272.2760, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(529.8779, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(132.7321, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(193.2873, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(132.9031, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(159.9095, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(118.8046, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(155.0196, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(87.5657, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(103.7380, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(192.4438, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(64.0326, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(68.9357, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(112.6656, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(83.7701, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(141.6772, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(142.3922, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(136.6180, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.8603, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.6705, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(85.5703, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.6537, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.1527, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.5194, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(70.1634, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(79.4770, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.1719, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.4146, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(151.5661, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(67.6226, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.4912, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(62.9508, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.3288, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.5542, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.0314, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(76.7637, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.9788, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.6259, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.7366, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.5333, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.0939, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.3411, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.9079, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(72.0741, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.1785, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(84.7353, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.5708, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.7908, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.7533, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.7095, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.0780, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.6822916666666666, 'recall': 0.3933933933933934, 'f1': 0.4990476190476191, 'number': 333}, 'P': {'precision': 0.44970414201183434, 'recall': 0.7676767676767676, 'f1': 0.5671641791044776, 'number': 297}, 'overall_precision': 0.5135908440629471, 'overall_recall': 0.5698412698412698, 'overall_f1': 0.5402558314522197, 'overall_accuracy': 0.653683746139787}
			------------EPOCH 19---------------
Loss:  tensor(92.1397, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(89.4876, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.8910, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(85.3165, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(92.2719, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.6338, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.2456, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.4721, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(65.8657, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(62.4704, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.5900, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.2881, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(100.4250, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(104.7917, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(66.4880, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(124.2699, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(61.5321, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.6841, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(116.6806, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(171.8162, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(147.3205, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(143.7175, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(207.5673, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(75.4111, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.4908, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(117.9743, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(66.5477, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(59.0865, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.1417, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.4300, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(71.9735, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.3564, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.8101, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(106.4280, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.5225, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.9416, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.9002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.7597, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.5857, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.3098, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.5285, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.4650, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.0885, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.0375, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.5950, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.2803, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.9685, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.8051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.7804, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.9759, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(62.6156, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.6569, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.3412, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.5613, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.4959, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.8752, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5438066465256798, 'recall': 0.5405405405405406, 'f1': 0.5421686746987951, 'number': 333}, 'P': {'precision': 0.5433526011560693, 'recall': 0.632996632996633, 'f1': 0.5847589424572317, 'number': 297}, 'overall_precision': 0.5435745937961596, 'overall_recall': 0.5841269841269842, 'overall_f1': 0.5631216526396329, 'overall_accuracy': 0.7226318774815655}
			------------EPOCH 20---------------
Loss:  tensor(94.7500, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(69.1803, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.1501, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(71.8464, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.9265, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.0657, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.2161, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.8376, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.8593, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.7310, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.5875, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.6263, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.0980, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.1439, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.1301, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(67.8236, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.3169, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.5246, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.3605, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.7115, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(102.1642, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(105.7645, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(82.0325, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.6885, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.0215, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.6822, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.8740, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.3960, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.4937, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.9389, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.8657, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.9587, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.9393, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.0355, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.0073, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.2143, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.1276, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.5031, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.1461, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.2631, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.5014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.7691, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.7896, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.7444, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.5131, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.4897, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.1676, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.5318, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.5806, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.5674, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.8315, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.5328, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.3069, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.2124, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.8295, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.0697, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5188172043010753, 'recall': 0.5795795795795796, 'f1': 0.5475177304964539, 'number': 333}, 'P': {'precision': 0.53125, 'recall': 0.5723905723905723, 'f1': 0.5510534846029174, 'number': 297}, 'overall_precision': 0.5245664739884393, 'overall_recall': 0.5761904761904761, 'overall_f1': 0.5491679273827533, 'overall_accuracy': 0.7102791958152139}
			------------EPOCH 21---------------
Loss:  tensor(78.9999, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(61.9561, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.9119, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.8601, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.8834, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.4934, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.7123, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.1845, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.1648, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.9443, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.5472, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.9670, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.9719, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.0199, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.6563, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.3829, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.3923, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.0713, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.9482, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.3446, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(91.2725, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(90.1531, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(66.3810, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.2259, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.1404, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.4794, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.8083, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.7843, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.5605, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.2591, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.0140, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.5907, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.9760, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.2363, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.7213, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.8213, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.5078, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.4343, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.1855, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.0041, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.0749, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.2521, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.2923, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.9943, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.4641, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.9661, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.1387, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.7847, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.3082, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.7323, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.4470, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.9860, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.9028, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.2488, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.5347, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.8165, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.49421965317919075, 'recall': 0.5135135135135135, 'f1': 0.503681885125184, 'number': 333}, 'P': {'precision': 0.5302593659942363, 'recall': 0.6195286195286195, 'f1': 0.5714285714285715, 'number': 297}, 'overall_precision': 0.5122655122655123, 'overall_recall': 0.5634920634920635, 'overall_f1': 0.5366591080876797, 'overall_accuracy': 0.7095229091825802}
			------------EPOCH 22---------------
Loss:  tensor(62.3115, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.8780, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.9538, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.4276, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.7676, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.7902, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.0769, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.1565, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.4496, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.4080, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.3542, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.0784, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.4868, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.0840, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.3687, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.3599, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.9183, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.8983, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.1045, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.3190, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(85.9803, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(83.8228, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.5493, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.7874, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.6105, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.5810, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.6724, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.6295, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.0263, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.4104, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.3295, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.6122, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.7234, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.7273, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.9168, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.5134, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.5252, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.5271, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.7058, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.4916, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.8625, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.5052, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.5129, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.1284, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.3325, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.1985, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.5594, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.3607, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.1419, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.4471, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.8851, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.9526, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.2691, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.4305, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.9580, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.8532, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5194029850746269, 'recall': 0.5225225225225225, 'f1': 0.5209580838323353, 'number': 333}, 'P': {'precision': 0.5492957746478874, 'recall': 0.6565656565656566, 'f1': 0.598159509202454, 'number': 297}, 'overall_precision': 0.5347826086956522, 'overall_recall': 0.5857142857142857, 'overall_f1': 0.5590909090909092, 'overall_accuracy': 0.7207411608999811}
			------------EPOCH 23---------------
Loss:  tensor(57.4863, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.0737, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.1001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.6446, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.9186, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.7258, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.5442, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.1544, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.2940, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.0769, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.1632, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.6061, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.5104, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.0732, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.7084, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.4835, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.7886, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.0431, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.6964, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.4863, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(82.0420, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(76.2827, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.1734, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.8311, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.0995, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.4953, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.8156, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.8466, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.3361, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.0570, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.7462, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.8357, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.0594, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.6380, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.1351, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.7067, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.7347, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.4909, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.3697, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.2470, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.4884, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.6732, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.8698, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.8382, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.2411, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.0611, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.2255, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.7405, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.0467, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.4859, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.8408, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.3096, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.3369, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.5208, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.3235, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.0174, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5308641975308642, 'recall': 0.5165165165165165, 'f1': 0.5235920852359208, 'number': 333}, 'P': {'precision': 0.5313351498637602, 'recall': 0.6565656565656566, 'f1': 0.5873493975903614, 'number': 297}, 'overall_precision': 0.5311143270622286, 'overall_recall': 0.5825396825396826, 'overall_f1': 0.5556396669190007, 'overall_accuracy': 0.7136824856620659}
			------------EPOCH 24---------------
Loss:  tensor(54.8528, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.6596, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.5976, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.5889, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.3793, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.5584, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.6923, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.1873, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.5834, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.4960, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.6388, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.9476, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.2434, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.5349, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.1746, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.1046, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.9268, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.0549, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.8827, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.5824, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(79.2346, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(70.4270, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.4560, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.1332, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.1216, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.0224, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.8560, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.3907, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.7913, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.5571, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.3470, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.2540, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.0334, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.1433, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.3847, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.5454, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.7467, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.4819, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.0706, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.7975, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.8821, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.4768, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.8577, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.4347, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.7068, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.8279, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.0340, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.0052, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.9647, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.7140, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.4556, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.4462, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.7605, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.4320, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.6736, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.3856, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5196374622356495, 'recall': 0.5165165165165165, 'f1': 0.5180722891566265, 'number': 333}, 'P': {'precision': 0.5357142857142857, 'recall': 0.6565656565656566, 'f1': 0.590015128593041, 'number': 297}, 'overall_precision': 0.5280575539568345, 'overall_recall': 0.5825396825396826, 'overall_f1': 0.5539622641509434, 'overall_accuracy': 0.7085145270057351}
			------------EPOCH 25---------------
Loss:  tensor(51.2988, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.8289, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.4047, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.6237, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.9453, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.4522, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.3399, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.1475, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.4281, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.8482, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.6617, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.2331, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.8098, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.5529, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.3577, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.5102, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.1637, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.1829, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.6998, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.9688, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(76.2533, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(65.4907, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.2022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.5504, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.1453, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.7811, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.3898, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.0411, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.3649, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.0860, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.2404, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.6799, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.5910, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.9487, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.0301, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.5818, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.8347, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.3286, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.1021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.2774, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.6218, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.2403, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.9300, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.1989, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.6898, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.4401, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.9411, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.3723, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.8762, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.9854, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.4479, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.7928, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.2812, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.5319, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.6042, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.9221, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.524390243902439, 'recall': 0.5165165165165165, 'f1': 0.5204236006051438, 'number': 333}, 'P': {'precision': 0.5309973045822103, 'recall': 0.6632996632996633, 'f1': 0.5898203592814371, 'number': 297}, 'overall_precision': 0.5278969957081545, 'overall_recall': 0.5857142857142857, 'overall_f1': 0.5553047404063206, 'overall_accuracy': 0.7240184029747274}
			------------EPOCH 26---------------
Loss:  tensor(47.9283, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.2606, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.9232, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.2050, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.8665, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.2525, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.1266, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.2713, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.3959, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.3486, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.7235, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.4880, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.0668, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.3501, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.1078, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.9441, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.4348, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.3782, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.7830, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.8567, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(73.2253, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(61.1336, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.6708, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.3203, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.3385, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.8468, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.4421, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.5230, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.0104, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.5606, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.9880, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.1384, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.1462, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.0269, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.8506, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.7843, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.9710, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.5092, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.3394, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.7929, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.9417, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.1861, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.1968, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.1646, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.9443, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.3180, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.1461, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.8125, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.2578, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.3873, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.9338, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.2344, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.8844, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.0526, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.7705, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.4638, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.534375, 'recall': 0.5135135135135135, 'f1': 0.5237366003062788, 'number': 333}, 'P': {'precision': 0.5349462365591398, 'recall': 0.67003367003367, 'f1': 0.5949177877428998, 'number': 297}, 'overall_precision': 0.5346820809248555, 'overall_recall': 0.5873015873015873, 'overall_f1': 0.5597579425113465, 'overall_accuracy': 0.7230100207978823}
			------------EPOCH 27---------------
Loss:  tensor(44.8975, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.9384, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.5246, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.2233, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.9531, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.2286, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.0392, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.5663, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.5552, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.9765, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.8944, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.8278, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.1692, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.5116, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.1345, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.5507, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.8490, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.7110, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.9096, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.1315, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(70.6521, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.3727, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.5881, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.3215, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.6194, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.0503, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.5930, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.9243, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.5823, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.0586, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.8432, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.6680, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.9325, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.3497, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.8738, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.0384, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.2463, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.8951, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.7264, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.5742, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.7616, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.2194, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.6799, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.2683, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.3366, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.3470, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.4767, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.3500, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.9132, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.8974, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.6508, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.7195, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.5365, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.9669, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.0987, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.0016, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.534375, 'recall': 0.5135135135135135, 'f1': 0.5237366003062788, 'number': 333}, 'P': {'precision': 0.5308310991957105, 'recall': 0.6666666666666666, 'f1': 0.591044776119403, 'number': 297}, 'overall_precision': 0.5324675324675324, 'overall_recall': 0.5857142857142857, 'overall_f1': 0.5578231292517006, 'overall_accuracy': 0.722820949139724}
			------------EPOCH 28---------------
Loss:  tensor(42.1725, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.7881, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.2016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.5320, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.1501, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.4852, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.1121, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.9739, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.8739, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.7651, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.1515, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.2328, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.3796, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.6572, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.3715, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.4181, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.5086, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.1273, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.0378, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.4437, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(68.4019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.0135, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.4049, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.4400, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.9628, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.3209, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.8348, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.4508, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.0970, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.5833, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.7308, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.2503, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.9838, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.8688, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.0445, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.2840, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.6117, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.3875, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.2163, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.6245, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.8633, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.3602, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.4409, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.4718, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.8403, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.4699, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.8823, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.0657, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.7575, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.4694, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.4730, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.2558, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.2295, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.1144, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.5485, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.5974, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5345911949685535, 'recall': 0.5105105105105106, 'f1': 0.5222734254992321, 'number': 333}, 'P': {'precision': 0.5197889182058048, 'recall': 0.6632996632996633, 'f1': 0.5828402366863906, 'number': 297}, 'overall_precision': 0.5265423242467718, 'overall_recall': 0.5825396825396826, 'overall_f1': 0.5531273549359458, 'overall_accuracy': 0.7243965462910443}
			------------EPOCH 29---------------
Loss:  tensor(39.7161, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.7486, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.9128, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.0846, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.4268, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.8498, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.3403, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.4588, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.2844, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.6920, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.5066, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.6968, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.2824, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.2818, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.7488, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.4738, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.4337, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.6025, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.1781, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.8134, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(66.3859, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.9884, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.9275, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.6384, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.3466, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.6045, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.1626, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.2169, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.5333, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.1515, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.5798, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.8693, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.2339, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.5381, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.3129, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.5042, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.0415, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.9329, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.7860, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.9000, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.1329, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.5987, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.4875, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.7721, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.4169, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.6784, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.3410, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.9451, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.7677, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.0864, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.3781, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.8332, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.9546, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.4278, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.0740, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.2260, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5345911949685535, 'recall': 0.5105105105105106, 'f1': 0.5222734254992321, 'number': 333}, 'P': {'precision': 0.5210526315789473, 'recall': 0.6666666666666666, 'f1': 0.5849335302806499, 'number': 297}, 'overall_precision': 0.5272206303724928, 'overall_recall': 0.5841269841269842, 'overall_f1': 0.5542168674698795, 'overall_accuracy': 0.7253419045818366}
			------------EPOCH 30---------------
Loss:  tensor(37.5236, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.7776, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.6263, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.7936, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.7521, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.1894, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.6644, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.9989, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.7558, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.7067, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.9423, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.2221, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.5645, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.0557, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.2019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.6542, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.5884, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.1377, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.3336, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.2416, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(64.5869, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.2762, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.9377, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.8883, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.7433, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.8916, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.5358, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.1759, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.8770, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.7740, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.3604, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.5159, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.6074, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.3369, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.6483, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.7358, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.5190, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.5223, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.4144, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.3291, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.5273, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.9305, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.7726, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.1699, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.0454, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.9661, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.8445, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.9468, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.9241, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.7453, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.3671, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.4462, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.7093, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.8606, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.6515, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.8850, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5297805642633229, 'recall': 0.5075075075075075, 'f1': 0.5184049079754601, 'number': 333}, 'P': {'precision': 0.5301837270341208, 'recall': 0.6801346801346801, 'f1': 0.5958702064896756, 'number': 297}, 'overall_precision': 0.53, 'overall_recall': 0.5888888888888889, 'overall_f1': 0.5578947368421053, 'overall_accuracy': 0.7217495430768261}
	Train size: 50 Test size: 50


		-------------RUN 1-----------
			------------EPOCH 1---------------
Loss:  tensor(1815.6614, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1759.6283, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1522.2327, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2563.4102, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2266.7354, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1916.7075, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1747.9448, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2400.8452, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1531.7529, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1825.4357, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2592.9995, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3146.3511, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3176.3413, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3007.8696, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2518.4941, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2708.4451, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2602.2166, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3134.9214, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2812.7366, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2385.7998, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2522.4636, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1591.4309, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2228.3789, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2522.3721, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1471.7714, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3570.7344, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2385.6045, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2232.9502, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(861.7288, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1104.1663, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1507.3137, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2531.9368, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1045.6666, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2071.1721, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'number': 771}, 'P': {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'number': 950}, 'overall_precision': 0.0, 'overall_recall': 0.0, 'overall_f1': 0.0, 'overall_accuracy': 0.2702681774772217}
			------------EPOCH 2---------------
Loss:  tensor(1349.3629, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1295.4287, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1096.9561, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1799.7609, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1655.5398, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1438.2700, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1329.6615, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1966.6605, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1218.2031, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1557.8279, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2068.7974, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2627.9404, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2713.8682, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2561.6663, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2033.1036, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2281.6638, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2316.5410, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2725.3608, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2486.7065, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2092.5972, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2172.9468, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1375.2596, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1951.4214, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2205.2759, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1329.7788, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3173.0105, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2089.9890, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1931.9030, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(754.6853, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(970.7266, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1318.1013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2182.5188, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(890.4729, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1842.5166, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.509090909090909, 'recall': 0.03631647211413749, 'f1': 0.06779661016949154, 'number': 771}, 'P': {'precision': 0.11221122112211221, 'recall': 0.07157894736842105, 'f1': 0.08740359897172235, 'number': 950}, 'overall_precision': 0.14523449319213314, 'overall_recall': 0.0557815223707147, 'overall_f1': 0.08060453400503778, 'overall_accuracy': 0.4446996876855174}
			------------EPOCH 3---------------
Loss:  tensor(1114.0620, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1069.3987, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(913.0391, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1486.8906, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1356.4498, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1189.2664, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1113.6891, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1730.2622, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(998.7235, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1380.2571, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1830.5707, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2317.0237, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2321.1655, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2226.5808, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1719.9451, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1874.3326, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1935.1206, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2311.3669, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2118.0532, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1718.8306, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1784.0884, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1110.0056, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1578.9802, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1921.0201, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1114.7799, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2687.4282, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1778.2030, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1607.0642, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(617.7719, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(819.0085, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1134.7788, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1871.9373, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(770.9413, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1634.3210, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3, 'recall': 0.07782101167315175, 'f1': 0.12358393408856849, 'number': 771}, 'P': {'precision': 0.3682277318640955, 'recall': 0.42210526315789476, 'f1': 0.3933300637567435, 'number': 950}, 'overall_precision': 0.35764158262218776, 'overall_recall': 0.26786751888436955, 'overall_f1': 0.30631229235880403, 'overall_accuracy': 0.5851895826342823}
			------------EPOCH 4---------------
Loss:  tensor(748.8484, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(711.4093, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(667.0781, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1116.2756, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1064.4033, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(970.5146, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(912.8094, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1363.9772, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(779.2649, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1146.5188, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1532.2927, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2024.4946, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1825.3280, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1787.5820, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1368.8674, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1353.5986, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1648.7953, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1937.2231, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1938.3486, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1458.0890, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1308.1202, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(815.3883, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1252.3904, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1650.9929, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(940.8926, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2230.8816, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1424.9537, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1271.1023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(514.2332, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(665.4930, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(957.2164, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1578.3694, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(662.8810, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1399.2761, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.28653295128939826, 'recall': 0.1297016861219196, 'f1': 0.17857142857142858, 'number': 771}, 'P': {'precision': 0.3981145757795504, 'recall': 0.5778947368421052, 'f1': 0.47144697294976384, 'number': 950}, 'overall_precision': 0.3755787037037037, 'overall_recall': 0.37710633352701917, 'overall_f1': 0.3763409683966367, 'overall_accuracy': 0.6405286116201636}
			------------EPOCH 5---------------
Loss:  tensor(483.3209, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(459.4268, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(506.7900, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(880.9208, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(790.2715, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(783.1479, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(744.2895, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1079.0271, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(621.2762, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(897.8256, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1268.3246, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1482.2173, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1464.3489, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1393.3505, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1067.8638, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1000.2850, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1400.8688, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1564.3750, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1608.7941, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1104.4041, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(938.8617, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(664.9935, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1109.6406, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1319.0011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(787.0527, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1909.5762, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1158.6879, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(989.0712, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(378.0583, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(479.5279, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(685.6519, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1126.2361, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(621.1661, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1324.3204, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.34584450402144773, 'recall': 0.16731517509727625, 'f1': 0.22552447552447555, 'number': 771}, 'P': {'precision': 0.31712062256809337, 'recall': 0.5147368421052632, 'f1': 0.39245585874799355, 'number': 950}, 'overall_precision': 0.3227154046997389, 'overall_recall': 0.3590935502614759, 'overall_f1': 0.33993399339934, 'overall_accuracy': 0.6182794311230416}
			------------EPOCH 6---------------
Loss:  tensor(513.2158, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(407.8905, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(486.3649, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1015.5527, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(759.5706, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(853.5951, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(931.1803, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1062.1672, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(620.1775, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1005.9846, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1229.5959, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2285.0701, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1625.9136, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1494.9187, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1002.7398, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(878.8573, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1041.8081, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1539.0768, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1495.0903, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1150.8667, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(891.2864, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(696.5983, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1148.5649, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1120.9688, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(936.4354, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3087.4556, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1397.1630, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1373.7346, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(449.0721, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(534.4061, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(741.7468, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1325.5288, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(623.7809, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1414.4457, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.298, 'recall': 0.5797665369649806, 'f1': 0.393659180977543, 'number': 771}, 'P': {'precision': 0.5870786516853933, 'recall': 0.22, 'f1': 0.3200612557427259, 'number': 950}, 'overall_precision': 0.35344827586206895, 'overall_recall': 0.38117373619988376, 'overall_f1': 0.3667878110148169, 'overall_accuracy': 0.5721033477015203}
			------------EPOCH 7---------------
Loss:  tensor(486.1966, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(461.8599, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(339.8843, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(710.5378, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(946.0521, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(916.6683, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1069.3816, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1723.5879, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(668.7648, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1045.8699, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1218.6094, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1694.5991, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1353.0479, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1320.3649, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(954.3394, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(831.6281, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1065.4722, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1705.7268, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1674.0876, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1322.7964, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1406.4553, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(709.1066, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1359.6017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1472.8108, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1040.4578, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2737.2134, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1716.4224, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1532.9255, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(401.9005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(609.8765, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(764.3751, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1288.3662, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(612.4344, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1378.9442, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.30414746543778803, 'recall': 0.08560311284046693, 'f1': 0.13360323886639675, 'number': 771}, 'P': {'precision': 0.4995773457311919, 'recall': 0.6221052631578947, 'f1': 0.5541490857946554, 'number': 950}, 'overall_precision': 0.4692857142857143, 'overall_recall': 0.3817547937245787, 'overall_f1': 0.42101890419737265, 'overall_accuracy': 0.6210154092352167}
			------------EPOCH 8---------------
Loss:  tensor(434.4971, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(441.8387, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(535.3354, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(865.6538, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(715.9825, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(745.7009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(656.4789, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(972.8071, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(492.3613, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(768.4782, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1005.6730, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1105.5654, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1231.9790, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1165.0120, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(860.1851, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(849.3588, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1082.7654, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1266.9614, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1181.1842, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(945.8066, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(913.5261, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(458.3270, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(923.1633, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1123.6562, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(606.6281, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1553.0626, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(857.0059, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(726.3993, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(271.4137, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(312.5834, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(423.2632, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(726.6367, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(379.5052, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(817.6511, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4629080118694362, 'recall': 0.4046692607003891, 'f1': 0.4318339100346021, 'number': 771}, 'P': {'precision': 0.5866261398176292, 'recall': 0.6094736842105263, 'f1': 0.5978316985028395, 'number': 950}, 'overall_precision': 0.5364238410596026, 'overall_recall': 0.5177222545031959, 'overall_f1': 0.5269071555292727, 'overall_accuracy': 0.7117414758795138}
			------------EPOCH 9---------------
Loss:  tensor(275.6794, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(250.2395, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(266.6807, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(382.3886, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(352.5479, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(473.6279, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(419.4452, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(693.3205, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(320.0855, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(579.1057, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(550.6657, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(653.6276, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(770.2378, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(673.1495, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(462.6047, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(437.3593, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(673.2137, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(866.2712, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(918.5189, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(577.6511, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(508.8942, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(279.5056, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(763.1832, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(649.8655, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(402.5726, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1249.1836, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(458.2448, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(450.2272, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(151.8775, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(199.1322, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(257.2603, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(474.4840, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(280.6595, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(569.8482, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.44039270687237025, 'recall': 0.4072632944228275, 'f1': 0.4231805929919138, 'number': 771}, 'P': {'precision': 0.5751633986928104, 'recall': 0.5557894736842105, 'f1': 0.5653104925053533, 'number': 950}, 'overall_precision': 0.5162477007970571, 'overall_recall': 0.48925043579314353, 'overall_f1': 0.5023866348448688, 'overall_accuracy': 0.7118705314508428}
			------------EPOCH 10---------------
Loss:  tensor(216.0896, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(174.7592, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(172.5312, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(246.1934, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(215.4517, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(365.3687, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(342.9155, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(492.9675, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(230.8816, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(427.0898, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(382.6980, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(500.0171, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(525.6768, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(410.5645, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(290.4033, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(277.2203, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(443.4697, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(538.7771, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(662.2740, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(359.5278, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(332.5751, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(216.9376, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(693.1655, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(417.4966, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(308.8256, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(828.6544, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(278.5730, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(303.4371, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(100.1766, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(167.4036, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(166.5071, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(261.5575, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(161.1700, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(363.7217, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4664179104477612, 'recall': 0.324254215304799, 'f1': 0.38255547054322875, 'number': 771}, 'P': {'precision': 0.5420475319926874, 'recall': 0.6242105263157894, 'f1': 0.5802348336594912, 'number': 950}, 'overall_precision': 0.5171779141104295, 'overall_recall': 0.4898314933178385, 'overall_f1': 0.5031333930170099, 'overall_accuracy': 0.7147613762486127}
			------------EPOCH 11---------------
Loss:  tensor(148.2201, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(140.1324, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(137.2575, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(186.1571, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(136.5920, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(248.2937, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(178.6328, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(328.6172, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(153.2188, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(274.3991, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(205.0226, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(278.8092, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(253.0542, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(253.3482, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(165.3533, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(181.9160, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(433.3051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(355.8848, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(423.0575, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(225.3813, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(209.4322, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(197.5030, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(524.9924, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(378.9056, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(188.4478, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(532.1529, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(158.6126, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(174.0419, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(72.0116, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(91.1263, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(125.1327, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(159.4594, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(115.5437, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(231.4542, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4868035190615836, 'recall': 0.21530479896238652, 'f1': 0.2985611510791367, 'number': 771}, 'P': {'precision': 0.5080882352941176, 'recall': 0.7273684210526316, 'f1': 0.5982683982683983, 'number': 950}, 'overall_precision': 0.5038212815990594, 'overall_recall': 0.4979662986635677, 'overall_f1': 0.5008766803039159, 'overall_accuracy': 0.6927961180084145}
			------------EPOCH 12---------------
Loss:  tensor(123.2126, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(120.9630, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(105.3220, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(143.2654, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(141.9613, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(271.1737, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(238.8641, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(575.2297, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(162.6486, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(301.6599, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(237.5585, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(318.1414, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(177.9161, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(154.1944, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(127.2532, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(133.0941, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(271.6987, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(200.1503, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(256.8917, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(134.2097, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(210.8021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(209.3945, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(435.9898, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(399.5288, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(352.8398, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(660.9092, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(311.7426, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(196.5048, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(99.9281, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(78.3248, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(196.0384, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(333.0089, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(136.6343, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(180.2295, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4808510638297872, 'recall': 0.4396887159533074, 'f1': 0.45934959349593496, 'number': 771}, 'P': {'precision': 0.584051724137931, 'recall': 0.5705263157894737, 'f1': 0.5772097976570819, 'number': 950}, 'overall_precision': 0.5394978567054501, 'overall_recall': 0.5119116792562464, 'overall_f1': 0.5253428741800835, 'overall_accuracy': 0.7202591435872286}
			------------EPOCH 13---------------
Loss:  tensor(112.7580, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(92.2853, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(70.9642, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(113.8092, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(85.9843, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(185.0354, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(144.7074, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(194.3304, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(86.0513, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(151.0966, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(122.8900, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(161.0401, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(269.6422, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(286.9452, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(198.2567, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(201.6245, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(315.9181, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(522.0611, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(607.7657, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(286.7393, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(164.8585, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(109.7652, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(692.4619, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(239.5923, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(131.0059, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(324.2737, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(78.5734, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(92.9612, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.2196, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.5941, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(77.8164, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(97.5459, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(94.8333, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(140.5484, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4125105663567202, 'recall': 0.6329442282749675, 'f1': 0.4994882292732856, 'number': 771}, 'P': {'precision': 0.6291390728476821, 'recall': 0.4, 'f1': 0.48906048906048905, 'number': 950}, 'overall_precision': 0.48573027420257414, 'overall_recall': 0.5043579314352121, 'overall_f1': 0.4948688711516534, 'overall_accuracy': 0.6883307952404305}
			------------EPOCH 14---------------
Loss:  tensor(111.2143, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(77.2594, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(61.9398, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(106.9730, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(185.5011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(328.3882, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(306.7027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(601.6454, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(204.0783, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(522.9791, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(302.6528, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(443.9574, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(488.5175, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(336.7911, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(171.1062, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(212.8175, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(200.3776, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(143.1084, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(169.9404, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(78.4949, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(101.5800, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(79.1404, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(233.6279, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(133.4552, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(199.5093, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(379.1591, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(148.3362, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(228.4187, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(66.1694, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(142.0996, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(120.9027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(204.5773, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(261.3294, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(340.6959, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5190839694656488, 'recall': 0.17639429312581065, 'f1': 0.2633107454017425, 'number': 771}, 'P': {'precision': 0.5164681149264191, 'recall': 0.7757894736842105, 'f1': 0.6201093815734119, 'number': 950}, 'overall_precision': 0.5168738898756661, 'overall_recall': 0.5072632190586868, 'overall_f1': 0.5120234604105572, 'overall_accuracy': 0.6914539400665927}
			------------EPOCH 15---------------
Loss:  tensor(190.2802, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(139.3111, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(307.6905, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(217.5650, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(84.6420, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(151.6390, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(103.2137, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(151.7638, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(71.0378, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(97.8459, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(100.8890, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(137.3809, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(169.1047, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(159.2251, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(109.2324, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(98.7354, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(174.4829, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(156.5507, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(149.7158, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(73.9278, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(82.4403, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(68.1357, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(164.2184, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(99.7646, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(85.4521, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(203.9117, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(59.5268, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(76.9235, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.7822, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.3200, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(70.8569, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(90.1547, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(79.7847, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(107.5506, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.42828282828282827, 'recall': 0.5499351491569391, 'f1': 0.4815445769449177, 'number': 771}, 'P': {'precision': 0.6289893617021277, 'recall': 0.4978947368421053, 'f1': 0.5558166862514688, 'number': 950}, 'overall_precision': 0.5149253731343284, 'overall_recall': 0.5212085996513655, 'overall_f1': 0.5180479353162, 'overall_accuracy': 0.7099088867666418}
			------------EPOCH 16---------------
Loss:  tensor(89.9044, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(64.1249, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.8503, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(72.9303, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.7859, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(76.3987, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(119.5458, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(88.6407, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(69.6316, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(96.0921, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(78.0066, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(107.2769, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(114.5113, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(109.7516, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(95.0461, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(101.8889, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(191.5226, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(136.4450, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(187.7420, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(80.3450, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(82.6327, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(86.3163, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(154.5739, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(110.6358, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(78.2866, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(176.7680, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.8859, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(79.8734, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.0135, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.6127, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.7232, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(79.9236, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.6382, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(67.4120, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.46153846153846156, 'recall': 0.4669260700389105, 'f1': 0.4642166344294004, 'number': 771}, 'P': {'precision': 0.5852842809364549, 'recall': 0.5526315789473685, 'f1': 0.5684894423389281, 'number': 950}, 'overall_precision': 0.5277280858676208, 'overall_recall': 0.5142359093550262, 'overall_f1': 0.5208946439081813, 'overall_accuracy': 0.7148388095914101}
			------------EPOCH 17---------------
Loss:  tensor(67.9230, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.5238, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.4734, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(65.3901, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.9463, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(67.4773, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.1891, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(71.9442, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.4940, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.1871, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.9625, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(81.1108, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(93.3510, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(99.4762, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(74.3376, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(77.8436, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(122.4322, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(90.2798, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(84.8475, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.2462, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.5837, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.7066, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(91.0696, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.5266, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.7100, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(133.5665, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.4819, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(68.6625, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.8555, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.2490, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.0924, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.5999, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.7216, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.8654, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5313531353135313, 'recall': 0.4176394293125811, 'f1': 0.46768336964415397, 'number': 771}, 'P': {'precision': 0.5705574912891986, 'recall': 0.6894736842105263, 'f1': 0.6244041944709247, 'number': 950}, 'overall_precision': 0.557012542759407, 'overall_recall': 0.567693201626961, 'overall_f1': 0.5623021582733813, 'overall_accuracy': 0.7330614562630668}
			------------EPOCH 18---------------
Loss:  tensor(55.6108, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.2466, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.3513, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.9598, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.1222, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.4743, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.6157, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.7060, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.3248, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.1047, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.3748, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(66.2758, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(70.6803, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(84.4697, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(67.9981, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(67.9455, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(108.2526, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(74.5284, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(80.4767, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.6321, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.8613, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.2758, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(67.6035, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.7173, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.6731, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(96.8529, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.7286, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.9308, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.8950, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.4472, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.0104, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.4930, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.9367, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.5532, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5068912710566615, 'recall': 0.4293125810635538, 'f1': 0.46488764044943814, 'number': 771}, 'P': {'precision': 0.5754545454545454, 'recall': 0.6663157894736842, 'f1': 0.6175609756097561, 'number': 950}, 'overall_precision': 0.5499144324015972, 'overall_recall': 0.5601394538059268, 'overall_f1': 0.5549798503166379, 'overall_accuracy': 0.7279250445241721}
			------------EPOCH 19---------------
Loss:  tensor(50.1272, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.4070, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.1469, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.9900, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.1753, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.3996, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.1431, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.9378, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.6356, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.9501, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.1243, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.4641, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(61.6201, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(79.6524, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.5757, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.9197, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(92.2463, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.7880, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(59.6674, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.3511, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.8345, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.2619, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.3566, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.2988, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.1520, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(72.7055, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.3582, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.5730, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.1921, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.9515, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.6557, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.0771, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.8366, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.5211, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.512743628185907, 'recall': 0.44357976653696496, 'f1': 0.47566063977746864, 'number': 771}, 'P': {'precision': 0.5726744186046512, 'recall': 0.6221052631578947, 'f1': 0.5963673057517658, 'number': 950}, 'overall_precision': 0.5491465567981165, 'overall_recall': 0.5421266705403835, 'overall_f1': 0.5456140350877193, 'overall_accuracy': 0.7235371550989856}
			------------EPOCH 20---------------
Loss:  tensor(44.4574, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.7657, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.8994, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.1060, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.8746, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.9330, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.9403, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.4178, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.4892, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.6038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.8173, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.8663, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.5778, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(78.4798, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.4737, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.8560, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(82.4361, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.1514, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.5989, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.5933, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.2086, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.3815, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.1931, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.6670, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.9379, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.4126, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.4633, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.2423, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.6097, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.3129, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.7958, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.9372, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.2392, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.5121, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.49786628733997157, 'recall': 0.45395590142671854, 'f1': 0.47489823609226595, 'number': 771}, 'P': {'precision': 0.5823529411764706, 'recall': 0.6252631578947369, 'f1': 0.6030456852791879, 'number': 950}, 'overall_precision': 0.5478816018572258, 'overall_recall': 0.5485183033120279, 'overall_f1': 0.5481997677119629, 'overall_accuracy': 0.726505433239553}
			------------EPOCH 21---------------
Loss:  tensor(40.9994, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.2708, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.4021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.8025, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.8537, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.9070, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.1639, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.4649, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.1235, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.5751, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.7624, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(59.7136, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.5382, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(75.7212, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.7225, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.6441, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(73.0016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.3586, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.0569, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.9124, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.2931, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.8241, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.9543, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.3716, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.7571, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.1872, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.1728, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.0567, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.4655, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.6943, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.7003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.5186, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.1006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.7046, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4992679355783309, 'recall': 0.44228274967574577, 'f1': 0.46905089408528194, 'number': 771}, 'P': {'precision': 0.5785920925747348, 'recall': 0.631578947368421, 'f1': 0.6039255158530448, 'number': 950}, 'overall_precision': 0.547093023255814, 'overall_recall': 0.546775130737943, 'overall_f1': 0.5469340308049986, 'overall_accuracy': 0.7266602999251478}
			------------EPOCH 22---------------
Loss:  tensor(40.4407, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.8255, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.4331, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.5329, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.5337, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.3536, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.7219, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.7978, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.4257, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.8931, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.6296, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.7778, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.6357, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(72.6072, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.4915, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.6399, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(67.3734, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.7807, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.9697, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.5184, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.3217, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.4850, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.6982, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.8644, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.2755, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.5788, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.5365, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.3921, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.8243, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.1607, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.2733, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.7650, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.6736, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.0466, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5029498525073747, 'recall': 0.44228274967574577, 'f1': 0.47066942719116633, 'number': 771}, 'P': {'precision': 0.5857699805068226, 'recall': 0.6326315789473684, 'f1': 0.6082995951417004, 'number': 950}, 'overall_precision': 0.5528169014084507, 'overall_recall': 0.547356188262638, 'overall_f1': 0.55007299270073, 'overall_accuracy': 0.7267893554964768}
			------------EPOCH 23---------------
Loss:  tensor(38.6663, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.8737, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.6457, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.3025, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.8253, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.0918, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.7458, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.4286, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.8963, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.3461, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.3116, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.5092, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.5774, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(74.4516, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.4289, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.3104, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.9531, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.9239, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.6208, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.3286, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.3427, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.3716, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.6344, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.8170, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.4208, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.9806, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.0784, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.9896, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.4179, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.7972, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.7717, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.6776, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.7467, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.6974, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5066273932253313, 'recall': 0.4461738002594034, 'f1': 0.47448275862068967, 'number': 771}, 'P': {'precision': 0.5866408518877058, 'recall': 0.6378947368421053, 'f1': 0.6111951588502269, 'number': 950}, 'overall_precision': 0.5549065420560748, 'overall_recall': 0.5520046484601976, 'overall_f1': 0.5534517914360618, 'overall_accuracy': 0.7281573445525643}
			------------EPOCH 24---------------
Loss:  tensor(36.6002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.3323, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.8841, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.2886, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.4819, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.7180, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.5993, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.9980, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.5718, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.0839, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.5963, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.8416, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.3722, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(72.5752, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.0296, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.6034, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.1390, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.3193, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.6694, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.4216, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.1462, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.6630, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.2804, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.6504, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.7975, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.4604, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.0537, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.1940, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.4110, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.8524, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.6265, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.3380, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.6299, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.2191, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5080763582966226, 'recall': 0.4487678339818418, 'f1': 0.47658402203856753, 'number': 771}, 'P': {'precision': 0.5902111324376199, 'recall': 0.6473684210526316, 'f1': 0.6174698795180722, 'number': 950}, 'overall_precision': 0.5577481137550784, 'overall_recall': 0.5583962812318419, 'overall_f1': 0.5580720092915215, 'overall_accuracy': 0.7290865446661332}
			------------EPOCH 25---------------
Loss:  tensor(35.0991, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.8975, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.2625, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.3051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.6836, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.7160, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.6497, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.5976, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.6733, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.4351, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.5116, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.0743, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.8393, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(71.7853, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.1967, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.6706, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.0121, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.9601, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.1754, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.4972, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.4676, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.2269, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.3514, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.2769, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.0989, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.3520, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.3868, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.8378, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.9228, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.8337, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.6269, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.3939, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.1402, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.2595, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.508100147275405, 'recall': 0.4474708171206226, 'f1': 0.47586206896551725, 'number': 771}, 'P': {'precision': 0.5775862068965517, 'recall': 0.6347368421052632, 'f1': 0.6048144433299899, 'number': 950}, 'overall_precision': 0.5502031340684852, 'overall_recall': 0.5508425334108077, 'overall_f1': 0.5505226480836238, 'overall_accuracy': 0.7281315334382985}
			------------EPOCH 26---------------
Loss:  tensor(33.4294, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.8583, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.0511, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.1887, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.2868, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.5383, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.1762, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.2190, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.8249, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.8540, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.7112, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.4148, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.5560, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(70.3681, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.4586, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.8428, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.5306, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.5459, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.0217, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.6220, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.1925, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.2144, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.9353, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.6091, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.9318, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.0311, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.8921, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.0145, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.5679, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.1078, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.8932, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.5468, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.7162, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.9267, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.49928469241773965, 'recall': 0.45265888456549935, 'f1': 0.4748299319727891, 'number': 771}, 'P': {'precision': 0.5808612440191387, 'recall': 0.6389473684210526, 'f1': 0.6085213032581454, 'number': 950}, 'overall_precision': 0.5481651376146789, 'overall_recall': 0.5554909936083672, 'overall_f1': 0.5518037518037516, 'overall_accuracy': 0.7272023333247296}
			------------EPOCH 27---------------
Loss:  tensor(32.1193, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.9994, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.5458, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.9879, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.2961, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.3334, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.5791, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.6277, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.1598, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.8251, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.4886, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.4927, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.8391, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(68.8232, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.2175, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.5981, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.1549, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.9239, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.2402, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.9059, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.5843, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.6022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.1954, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.1541, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.8196, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.1633, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.8472, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.1799, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.9760, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.0838, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.3081, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.0811, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.6503, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.9876, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.49445983379501385, 'recall': 0.46303501945525294, 'f1': 0.47823174815807107, 'number': 771}, 'P': {'precision': 0.5831739961759083, 'recall': 0.6421052631578947, 'f1': 0.6112224448897797, 'number': 950}, 'overall_precision': 0.5469457013574661, 'overall_recall': 0.5618826263800116, 'overall_f1': 0.5543135568930926, 'overall_accuracy': 0.7278992334099063}
			------------EPOCH 28---------------
Loss:  tensor(31.0505, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.3954, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.8955, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.1423, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.3905, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.3783, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.0893, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.8357, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.2277, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.5154, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.7708, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.3749, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.0472, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.8687, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.8488, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.8367, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.7779, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.7081, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.2345, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.6599, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.2601, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.5988, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.4656, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.1507, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.1294, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.0768, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.3861, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.1393, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.1642, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.0972, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.3130, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.7989, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.3936, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.2524, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4958100558659218, 'recall': 0.4604409857328145, 'f1': 0.47747141896435774, 'number': 771}, 'P': {'precision': 0.581573896353167, 'recall': 0.6378947368421053, 'f1': 0.6084337349397592, 'number': 950}, 'overall_precision': 0.5466439135381115, 'overall_recall': 0.5583962812318419, 'overall_f1': 0.5524576027594137, 'overall_accuracy': 0.7283122112381591}
			------------EPOCH 29---------------
Loss:  tensor(29.9297, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.6970, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.3210, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.7013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.4892, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.3802, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.5272, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.9101, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.8349, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.8208, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.7199, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.8007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.1093, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.5756, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.6246, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.9705, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.3748, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.2066, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.4019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.5676, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.2814, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.0256, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.6832, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.9003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.0165, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.9793, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.8909, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.3278, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.4218, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.2848, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.5485, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.9339, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.9904, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.3187, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.502086230876217, 'recall': 0.4682230869001297, 'f1': 0.4845637583892617, 'number': 771}, 'P': {'precision': 0.576222435282838, 'recall': 0.6326315789473684, 'f1': 0.6031108881083794, 'number': 950}, 'overall_precision': 0.5459704880817253, 'overall_recall': 0.5589773387565369, 'overall_f1': 0.552397358598909, 'overall_accuracy': 0.7280541000955011}
			------------EPOCH 30---------------
Loss:  tensor(28.8406, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.0871, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.9437, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.3028, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.0131, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.2804, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.6019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.1874, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.2159, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.0046, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.9266, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.6995, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.7837, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.0155, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.0413, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.9460, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.8513, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.7046, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.3547, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.8121, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.0510, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.3672, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.6518, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.0182, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.3005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.3421, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.4628, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.5045, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.9007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.7888, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.0940, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.2536, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.7530, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.0322, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.502770083102493, 'recall': 0.4708171206225681, 'f1': 0.48626925653047554, 'number': 771}, 'P': {'precision': 0.5798076923076924, 'recall': 0.6347368421052632, 'f1': 0.6060301507537689, 'number': 950}, 'overall_precision': 0.5482406356413166, 'overall_recall': 0.5613015688553167, 'overall_f1': 0.5546942291128337, 'overall_accuracy': 0.7270732777534006}


		-------------RUN 2-----------
			------------EPOCH 1---------------
Loss:  tensor(1880.8829, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2005.0388, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2507.7441, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1538.8752, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1007.4410, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1695.6353, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1485.4756, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2679.2798, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3036.3672, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1108.3118, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2623.7563, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2635.0833, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2831.1941, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1116.2418, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1766.0516, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2855.9001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(958.0623, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1380.8398, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2478.4138, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2549.8799, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1620.4911, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(452.6688, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1027.5901, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1318.0256, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(809.0874, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1013.9996, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1141.5375, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1741.6973, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(993.4772, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(923.0016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2682.5430, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2652.5425, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2595.8237, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1109.1509, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1459.1405, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1863.1565, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.054878048780487805, 'recall': 0.023017902813299233, 'f1': 0.032432432432432434, 'number': 782}, 'P': {'precision': 0.09437751004016064, 'recall': 0.149364406779661, 'f1': 0.11566858080393765, 'number': 944}, 'overall_precision': 0.08726673984632272, 'overall_recall': 0.0921205098493627, 'overall_f1': 0.08962795941375422, 'overall_accuracy': 0.5433483998455493}
			------------EPOCH 2---------------
Loss:  tensor(1227.2456, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1469.5679, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1876., device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1149.9380, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(735.7631, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1343.8235, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1195.6179, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2023.4429, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2501.3003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(899.7175, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2278.9863, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2089.0981, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2400.7129, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(899.1353, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1365.4033, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2402.9766, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(788.4925, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1153.1780, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2132.8655, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2193.5635, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1400.4321, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(381.9431, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(878.8145, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1076.5449, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(692.2670, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(855.9238, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(951.4917, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1481.7617, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(863.7007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(764.8675, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2226.9021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2296.9070, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2190.7014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(975.7234, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1232.2288, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1536.0332, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.21711899791231734, 'recall': 0.1329923273657289, 'f1': 0.16494845360824745, 'number': 782}, 'P': {'precision': 0.2669392523364486, 'recall': 0.4841101694915254, 'f1': 0.3441265060240964, 'number': 944}, 'overall_precision': 0.25604746691008673, 'overall_recall': 0.3250289687137891, 'overall_f1': 0.28644370691856014, 'overall_accuracy': 0.6076725645626547}
			------------EPOCH 3---------------
Loss:  tensor(996.3400, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1144.3132, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1551.7034, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(988.4758, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(597.3254, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1089.3826, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(959.5245, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1689.3871, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2094.2524, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(762.9300, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1742.1998, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1828.0524, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1995.7769, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(780.7776, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1141.8954, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2128.0977, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(659.3940, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1018.2041, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1870.5374, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2018.8258, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1023.4045, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(305.4757, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(726.7891, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(863.0068, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(571.6280, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(732.3973, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(789.7037, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1241.0234, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(801.9481, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(632.9385, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2050.2976, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2183.9790, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2006.3519, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(868.5076, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1089.1117, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1348.8284, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.31223628691983124, 'recall': 0.28388746803069054, 'f1': 0.2973878097789685, 'number': 782}, 'P': {'precision': 0.36859838274932616, 'recall': 0.5794491525423728, 'f1': 0.4505766062602965, 'number': 944}, 'overall_precision': 0.35034168564920276, 'overall_recall': 0.4455388180764774, 'overall_f1': 0.3922468757969905, 'overall_accuracy': 0.6270470393167829}
			------------EPOCH 4---------------
Loss:  tensor(844.4551, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1023.8146, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1370.6833, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(890.8862, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(512.9033, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(909.1385, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(850.3451, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1512.4415, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1671.3215, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(683.9265, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1530.6519, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1616.2678, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1761.1802, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(747.2716, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(797.3051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1506.1240, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(563.3386, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(904.9614, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1737.7852, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1780.1687, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(780.7761, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(226.4285, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(726.8817, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(794.9881, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(455.8104, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(640.6577, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(776.1168, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1080.5803, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(643.0616, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(487.8307, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1500.3301, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1746.1371, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1419.5328, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(742.9279, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(793.4030, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(981.1938, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.26480836236933797, 'recall': 0.3887468030690537, 'f1': 0.3150259067357513, 'number': 782}, 'P': {'precision': 0.35195530726256985, 'recall': 0.4004237288135593, 'f1': 0.3746283448959366, 'number': 944}, 'overall_precision': 0.3069306930693069, 'overall_recall': 0.3951332560834299, 'overall_f1': 0.3454913880445796, 'overall_accuracy': 0.6068321711676925}
			------------EPOCH 5---------------
Loss:  tensor(787.1584, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(987.6511, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1234.9424, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(895.4001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(413.9475, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(802.3516, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(832.1528, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1315.0673, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1272.6055, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(582.4707, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1524.4526, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1420.9348, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1489.4817, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(633.5768, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(452.0046, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1046.8083, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(428.9044, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(644.4230, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1201.9885, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1238.1846, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(537.0001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(180.9558, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(631.6045, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(624.5456, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(368.4516, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(535.1270, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(759.8938, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(971.2181, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(554.2058, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(418.4287, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1179.0994, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1529.7297, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(990.0468, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(654.6810, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(656.9441, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(807.8265, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.34069400630914826, 'recall': 0.4143222506393862, 'f1': 0.3739180611656088, 'number': 782}, 'P': {'precision': 0.407773851590106, 'recall': 0.611228813559322, 'f1': 0.4891903348876643, 'number': 944}, 'overall_precision': 0.3808114961961116, 'overall_recall': 0.5220162224797219, 'overall_f1': 0.44037145650048876, 'overall_accuracy': 0.6600268017352988}
			------------EPOCH 6---------------
Loss:  tensor(605.1512, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(676.6311, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(986.2452, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(633.7347, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(346.6519, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(699.9641, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(673.3815, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(884.9706, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1021.6548, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(489.7188, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1009.9914, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1212.1923, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1060.8477, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(545.0052, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(410.0584, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(871.4047, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(312.9642, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(512.5256, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(979.7158, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1071.6211, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(464.3790, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(150.5972, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(511.9750, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(384.9087, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(244.1172, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(298.7639, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(446.8917, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(707.1920, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(444.6256, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(281.6195, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1132.3827, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1379.9829, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(709.0936, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(501.4576, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(489.4082, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(536.1462, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.2530765115034778, 'recall': 0.6048593350383632, 'f1': 0.35684647302904565, 'number': 782}, 'P': {'precision': 0.38190954773869346, 'recall': 0.08050847457627118, 'f1': 0.13298337707786526, 'number': 944}, 'overall_precision': 0.26547388781431336, 'overall_recall': 0.31807647740440326, 'overall_f1': 0.28940432261465476, 'overall_accuracy': 0.4838621754832262}
			------------EPOCH 7---------------
Loss:  tensor(921.6313, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1058.3580, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1313.3831, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(671.2509, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(299.4783, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(556.3374, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(541.7809, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(964.2744, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(649.2995, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(296.3227, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(371.5209, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(851.2681, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(747.1321, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(458.1673, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(235.8831, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(649.0015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(257.3625, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(467.2095, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(831.4171, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1058.7520, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(403.3963, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(122.0870, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(532.3806, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(374.6711, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(265.1217, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(293.0421, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(464.7010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(837.8024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(475.3633, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(329.0497, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1033.9603, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1072.9794, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(682.0490, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(474.0323, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(394.0408, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(458.9674, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3093278463648834, 'recall': 0.5767263427109974, 'f1': 0.40267857142857144, 'number': 782}, 'P': {'precision': 0.46143250688705234, 'recall': 0.3548728813559322, 'f1': 0.4011976047904192, 'number': 944}, 'overall_precision': 0.3598901098901099, 'overall_recall': 0.45538818076477405, 'overall_f1': 0.4020460358056266, 'overall_accuracy': 0.6194153587571263}
			------------EPOCH 8---------------
Loss:  tensor(414.0386, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(398.6144, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(665.0634, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(384.1270, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(238.3220, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(451.4600, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(435.8484, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(614.1380, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1144.7272, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(444.4774, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(665.2816, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1568.7152, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1226.8850, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(597.5237, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(447.1577, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(898.1156, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(266.7535, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(470.0286, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(776.1389, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1042.4719, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(214.8854, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(97.1761, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(374.0288, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(229.4400, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(274.6675, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(231.7224, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(423.4423, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(892.2376, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(540.3558, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(372.1320, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1593.0564, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1824.3970, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(999.7750, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(705.8392, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(571.4579, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(792.0802, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5, 'recall': 0.18414322250639387, 'f1': 0.2691588785046729, 'number': 782}, 'P': {'precision': 0.36853002070393376, 'recall': 0.7542372881355932, 'f1': 0.4951321279554938, 'number': 944}, 'overall_precision': 0.3855855855855856, 'overall_recall': 0.4959443800695249, 'overall_f1': 0.4338570704510897, 'overall_accuracy': 0.6211415722170486}
			------------EPOCH 9---------------
Loss:  tensor(736.6399, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(633.7723, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1545.3094, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(758.1850, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(252.0377, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(777.0455, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(670.5574, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(875.2955, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(773.7235, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(317.6696, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(596.4497, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(951.5320, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(796.4326, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(413.4495, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(345.1138, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(661.1201, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(317.3481, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(527.9849, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(769.8111, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(917.8764, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(786.9312, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(291.3617, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(568.9493, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(461.9765, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(357.0574, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(383.5412, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(538.1561, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(797.8793, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(466.9868, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(278.1255, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(815.0840, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1139.8917, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(632.6101, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(370.8638, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(443.1233, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(419.3455, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.26654240447343897, 'recall': 0.3657289002557545, 'f1': 0.308355795148248, 'number': 782}, 'P': {'precision': 0.38014311270125223, 'recall': 0.4502118644067797, 'f1': 0.4122211445198837, 'number': 944}, 'overall_precision': 0.32450935645823825, 'overall_recall': 0.4119351100811124, 'overall_f1': 0.363032933367373, 'overall_accuracy': 0.6802643832193881}
			------------EPOCH 10---------------
Loss:  tensor(451.4263, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(363.1389, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(669.7855, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(349.1680, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(218.8652, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(429.7035, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(558.6223, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(503.7177, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(551.4604, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(303.0201, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(276.7747, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(700.8419, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(540.2612, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(356.1734, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(249.9347, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(531.1914, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(120.6730, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(243.6671, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(356.6292, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(476.2925, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(193.8326, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(65.2298, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(283.4563, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(155.3850, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(123.5476, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(136.6813, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(233.6828, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(347.5127, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(227.5968, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(131.9922, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(547.1458, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(557.8613, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(351.5503, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(228.2470, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(238.2245, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(240.4709, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.44744744744744747, 'recall': 0.38107416879795397, 'f1': 0.4116022099447514, 'number': 782}, 'P': {'precision': 0.42915531335149865, 'recall': 0.6673728813559322, 'f1': 0.5223880597014926, 'number': 944}, 'overall_precision': 0.43486410496719774, 'overall_recall': 0.5376593279258401, 'overall_f1': 0.48082901554404145, 'overall_accuracy': 0.683489676789243}
			------------EPOCH 11---------------
Loss:  tensor(274.8322, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(219.5359, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(527.9826, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(247.3776, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(103.9510, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(341.7543, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(304.2766, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(259.6520, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(347.0692, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(152.8368, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(101.5462, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(299.5522, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(269.8876, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(263.5127, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(124.5595, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(273.7722, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(65.0407, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(137.4677, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(175.1643, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(263.9129, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(89.6345, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.8730, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(208.6537, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(109.7250, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(70.7492, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(100.1405, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(159.5601, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(269.5838, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(146.8790, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(93.3584, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(382.2330, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(318.1855, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(151.0779, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(135.5513, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(169.2288, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(149.5569, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.39298669891172916, 'recall': 0.4156010230179028, 'f1': 0.4039776258545681, 'number': 782}, 'P': {'precision': 0.45361702127659576, 'recall': 0.5646186440677966, 'f1': 0.5030674846625768, 'number': 944}, 'overall_precision': 0.42857142857142855, 'overall_recall': 0.49710312862108924, 'overall_f1': 0.46030042918454933, 'overall_accuracy': 0.7058168850932383}
			------------EPOCH 12---------------
Loss:  tensor(200.5409, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(141.3638, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(292.6335, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(188.3470, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(84.4309, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(165.8092, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(266.0667, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(109.0228, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(225.5873, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(121.5353, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.4805, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(130.9141, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(169.8358, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(181.8110, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.1304, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(152.6943, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.8644, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(89.5388, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(103.7690, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(149.9469, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.3425, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.8704, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(122.6227, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(68.1448, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.4891, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.7552, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(94.7106, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(145.2527, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(88.2749, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.5684, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(292.0327, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(187.7060, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(96.5810, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(92.3766, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(107.8803, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(94.1589, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.43443557582668185, 'recall': 0.48721227621483376, 'f1': 0.45931283905967446, 'number': 782}, 'P': {'precision': 0.48801916932907347, 'recall': 0.6472457627118644, 'f1': 0.5564663023679417, 'number': 944}, 'overall_precision': 0.4659464537341475, 'overall_recall': 0.574739281575898, 'overall_f1': 0.514656290531777, 'overall_accuracy': 0.6870329570490835}
			------------EPOCH 13---------------
Loss:  tensor(88.2838, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(72.6478, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(172.5185, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(101.0830, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.8347, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(120.4019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(90.4903, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(77.7641, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(174.9555, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(76.8136, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.8214, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(104.1628, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(115.6158, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(134.9033, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.2952, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(100.0835, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.3567, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(65.4483, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(70.2615, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(92.0430, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.7258, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.4104, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(75.2327, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.0877, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.3611, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.3906, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.1905, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(113.3223, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.2854, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.0160, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(215.2600, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(113.3634, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(64.7301, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(64.8419, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(70.5331, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.6915, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.43380614657210403, 'recall': 0.46930946291560105, 'f1': 0.4508599508599509, 'number': 782}, 'P': {'precision': 0.4902459711620017, 'recall': 0.6122881355932204, 'f1': 0.5445124823363164, 'number': 944}, 'overall_precision': 0.4666666666666667, 'overall_recall': 0.5475086906141368, 'overall_f1': 0.5038656358304453, 'overall_accuracy': 0.7062257251232198}
			------------EPOCH 14---------------
Loss:  tensor(65.3260, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.2511, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(98.0213, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(72.7595, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.3162, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(78.2132, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.3972, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.6118, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(81.7149, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.3960, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.5983, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(64.1860, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(84.9395, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(96.2305, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.2531, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(64.2806, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.8709, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.8576, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.3358, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(65.0471, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.1899, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.0083, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.8254, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.0761, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.0039, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.1183, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.5143, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(89.2336, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.6044, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.5806, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(163.6764, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(73.8802, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.3475, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.3564, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.8092, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.4735, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4664310954063604, 'recall': 0.5063938618925832, 'f1': 0.4855916615573268, 'number': 782}, 'P': {'precision': 0.506677796327212, 'recall': 0.6430084745762712, 'f1': 0.5667600373482726, 'number': 944}, 'overall_precision': 0.4899853444064485, 'overall_recall': 0.5811123986095017, 'overall_f1': 0.5316724092234297, 'overall_accuracy': 0.7075658118881596}
			------------EPOCH 15---------------
Loss:  tensor(45.6804, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.3129, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(80.0450, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.4515, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.5465, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.8476, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.6861, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.0167, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(75.4713, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.0123, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.3987, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.0009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(83.7291, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(81.5708, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.6103, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.3800, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.8397, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.6623, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.9195, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.6271, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.0923, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.3184, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.7654, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.1566, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.6810, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.6104, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.5092, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(72.7435, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.2612, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.3638, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(116.5882, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.7460, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.5159, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.7491, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.5954, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.9534, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.44075829383886256, 'recall': 0.47570332480818417, 'f1': 0.45756457564575653, 'number': 782}, 'P': {'precision': 0.49794238683127573, 'recall': 0.6408898305084746, 'f1': 0.5604446503010654, 'number': 944}, 'overall_precision': 0.47450218552695483, 'overall_recall': 0.5660486674391657, 'overall_f1': 0.5162483487450462, 'overall_accuracy': 0.6998205646535081}
			------------EPOCH 16---------------
Loss:  tensor(34.4344, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.2562, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(62.8495, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.7932, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.9882, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.8211, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.8008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.1704, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(65.4574, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.3344, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.6228, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.1819, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(82.3526, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(64.0493, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.4566, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.2006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.1977, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.4577, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.4438, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.9694, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.5130, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.5831, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.3223, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.0089, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.0621, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.3704, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.9900, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.4156, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.3380, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.4787, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(83.3459, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.1813, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.0499, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.3488, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.1614, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.5081, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.43861740166865315, 'recall': 0.47058823529411764, 'f1': 0.4540407156076496, 'number': 782}, 'P': {'precision': 0.49503311258278143, 'recall': 0.6334745762711864, 'f1': 0.5557620817843866, 'number': 944}, 'overall_precision': 0.47191011235955055, 'overall_recall': 0.5596755504055619, 'overall_f1': 0.5120593692022263, 'overall_accuracy': 0.6999795579985009}
			------------EPOCH 17---------------
Loss:  tensor(25.7866, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.4436, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.9636, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.7801, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.6396, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.5102, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.3696, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.8310, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.9925, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.2271, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.8562, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.6597, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.4822, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.9897, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.0707, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.8054, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.6070, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.5710, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.0382, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.1971, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.5603, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.7669, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.9111, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.3372, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.8341, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.7346, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.5339, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.3016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.2376, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.7163, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(65.8647, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.6132, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.6414, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.9120, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.8733, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.1358, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.45717884130982367, 'recall': 0.4641943734015345, 'f1': 0.4606598984771573, 'number': 782}, 'P': {'precision': 0.48186946011281223, 'recall': 0.6334745762711864, 'f1': 0.5473684210526316, 'number': 944}, 'overall_precision': 0.4722358722358722, 'overall_recall': 0.5567786790266512, 'overall_f1': 0.5110342993884605, 'overall_accuracy': 0.6981852045335817}
			------------EPOCH 18---------------
Loss:  tensor(19.6355, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.3644, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.8546, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.1007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.6569, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.2803, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.8186, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.8375, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.4009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.0436, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.1375, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.6703, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.7832, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.8084, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.2218, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.7463, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.5758, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.3449, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.3436, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.3742, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.0566, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.0928, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.9366, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.8654, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.9449, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.6435, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.4325, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.9503, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.3700, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.1503, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.9492, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.2943, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.3172, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.9591, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.2677, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.8700, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.440925700365408, 'recall': 0.4629156010230179, 'f1': 0.45165315034310666, 'number': 782}, 'P': {'precision': 0.48663967611336034, 'recall': 0.6366525423728814, 'f1': 0.5516291877007802, 'number': 944}, 'overall_precision': 0.4683852140077821, 'overall_recall': 0.5579374275782155, 'overall_f1': 0.5092543627710207, 'overall_accuracy': 0.6972766711336226}
			------------EPOCH 19---------------
Loss:  tensor(16.2060, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.7300, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.8544, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.7153, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.3272, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.4725, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.3799, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.3808, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.6228, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.9598, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.4097, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.8355, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.1951, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.2170, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.2874, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.2408, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.2776, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.9714, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.1509, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.0860, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.7071, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.4849, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.3091, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.4190, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.8621, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.9492, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.4238, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.7213, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.0425, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.4658, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.2204, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.7911, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.7767, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.7062, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.7613, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.8371, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.44870210135970334, 'recall': 0.4641943734015345, 'f1': 0.45631678189817726, 'number': 782}, 'P': {'precision': 0.4840764331210191, 'recall': 0.6440677966101694, 'f1': 0.5527272727272727, 'number': 944}, 'overall_precision': 0.47021791767554477, 'overall_recall': 0.5625724217844728, 'overall_f1': 0.512265892904247, 'overall_accuracy': 0.6981170645285847}
			------------EPOCH 20---------------
Loss:  tensor(14.0227, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.1414, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.6660, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.7949, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.7096, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.2750, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.8220, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.9006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.6579, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.9869, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.4904, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.2747, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.5184, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.9343, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.0651, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.5191, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.4174, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.8730, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.6219, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.7051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.6886, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.0075, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.5603, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.9832, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.9270, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.9443, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.3238, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.0831, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.8158, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.3688, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.0458, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.2761, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.7698, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.3304, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.9181, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.2157, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4351851851851852, 'recall': 0.48081841432225064, 'f1': 0.456865127582017, 'number': 782}, 'P': {'precision': 0.48976248976248976, 'recall': 0.6334745762711864, 'f1': 0.5524249422632794, 'number': 944}, 'overall_precision': 0.4671462829736211, 'overall_recall': 0.5643105446118193, 'overall_f1': 0.5111519286276568, 'overall_accuracy': 0.700956231403457}
			------------EPOCH 21---------------
Loss:  tensor(11.9018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.0389, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.0254, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.2284, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.2365, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.1932, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.5062, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.8346, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.6680, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.9085, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.5395, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.8511, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.0822, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.5941, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.9872, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.4457, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.5464, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.2373, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.5444, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.5075, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.8937, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.5809, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.3529, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.9110, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.8720, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.9751, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.3567, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.6911, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.6790, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.5820, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.7641, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.5457, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.8890, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.4339, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.2916, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.2104, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.44244604316546765, 'recall': 0.4718670076726343, 'f1': 0.4566831683168317, 'number': 782}, 'P': {'precision': 0.48589846897663175, 'recall': 0.638771186440678, 'f1': 0.5519450800915332, 'number': 944}, 'overall_precision': 0.46843373493975904, 'overall_recall': 0.5631517960602549, 'overall_f1': 0.5114443567482241, 'overall_accuracy': 0.7012515047584437}
			------------EPOCH 22---------------
Loss:  tensor(9.9094, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.2324, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.8284, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.1958, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.0999, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.4770, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.9865, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.0746, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.5375, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.7231, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.0924, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.5363, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.9065, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.0756, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.0549, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.5993, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.7486, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.5232, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.7218, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.8301, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.1321, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.1020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.9086, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.0211, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.2323, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.1819, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.8138, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.0045, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.8167, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.0253, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.7545, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.6218, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.4295, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.8214, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.2499, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.8921, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4410011918951132, 'recall': 0.4731457800511509, 'f1': 0.4565083281924738, 'number': 782}, 'P': {'precision': 0.4882970137207425, 'recall': 0.6408898305084746, 'f1': 0.5542830966559781, 'number': 944}, 'overall_precision': 0.4692011549566891, 'overall_recall': 0.5648899188876014, 'overall_f1': 0.5126182965299685, 'overall_accuracy': 0.7034319849183456}
			------------EPOCH 23---------------
Loss:  tensor(7.9255, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.7025, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.1347, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.8772, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.0421, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.7304, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.8964, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.9734, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.0621, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.6838, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.5272, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.7436, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.0217, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.3677, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.2645, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.3560, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.1695, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.1646, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.1347, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.5029, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.5359, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.6607, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.7577, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.7559, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.0478, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.6068, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.7571, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.5782, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.2281, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.4732, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.3629, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.8811, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.3478, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.4544, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.3792, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.8986, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4439140811455847, 'recall': 0.47570332480818417, 'f1': 0.45925925925925926, 'number': 782}, 'P': {'precision': 0.4870967741935484, 'recall': 0.6398305084745762, 'f1': 0.553113553113553, 'number': 944}, 'overall_precision': 0.46968238691049086, 'overall_recall': 0.5654692931633836, 'overall_f1': 0.5131440588853837, 'overall_accuracy': 0.7025007381833874}
			------------EPOCH 24---------------
Loss:  tensor(6.4845, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.2110, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.9337, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.4440, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.3595, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.9940, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.1477, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.8779, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.7708, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.3951, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.8800, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.3957, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.3788, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.8462, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.5550, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.1759, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.7075, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.3120, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.3094, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.5387, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.0177, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.3752, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.1862, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.8877, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.1683, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.0258, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.9908, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.4742, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.7488, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.9449, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.3220, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.2422, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.4403, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.2897, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.5476, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.1236, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.44591346153846156, 'recall': 0.4744245524296675, 'f1': 0.459727385377943, 'number': 782}, 'P': {'precision': 0.4915254237288136, 'recall': 0.6451271186440678, 'f1': 0.5579477782867612, 'number': 944}, 'overall_precision': 0.4732013520038629, 'overall_recall': 0.5677867902665121, 'overall_f1': 0.5161969976297076, 'overall_accuracy': 0.7016149181184274}
			------------EPOCH 25---------------
Loss:  tensor(5.4182, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.0905, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.9819, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.3059, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.8397, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.3742, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.5595, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.0392, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.5361, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.3097, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.3622, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.2806, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.0196, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.5862, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.0370, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.2238, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.2213, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.6078, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.6227, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.7103, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.5758, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.0753, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.0192, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.1343, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.4892, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.5439, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.3250, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.9527, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.3053, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.5015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.6704, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.0323, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.6344, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.2647, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.8813, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.5388, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4520383693045564, 'recall': 0.4820971867007673, 'f1': 0.46658415841584155, 'number': 782}, 'P': {'precision': 0.4906731549067315, 'recall': 0.6408898305084746, 'f1': 0.5558107487367938, 'number': 944}, 'overall_precision': 0.47508466376390907, 'overall_recall': 0.5689455388180765, 'overall_f1': 0.5177959398892698, 'overall_accuracy': 0.7031594248983578}
			------------EPOCH 26---------------
Loss:  tensor(4.6054, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.2399, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.0963, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.2450, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.3984, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.9131, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.0336, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.3774, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.4149, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.4626, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.3436, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.4243, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.0613, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.6087, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.5812, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.3991, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.7559, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.9462, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.1160, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.9688, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.2112, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.7622, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.1446, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.5111, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.9913, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.1420, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.7655, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.5577, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.9511, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.0902, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.1876, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.9728, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.6007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.3683, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.2903, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.0463, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.45149700598802395, 'recall': 0.4820971867007673, 'f1': 0.4662956091527521, 'number': 782}, 'P': {'precision': 0.49227013832384053, 'recall': 0.6408898305084746, 'f1': 0.5568338702254948, 'number': 944}, 'overall_precision': 0.4757751937984496, 'overall_recall': 0.5689455388180765, 'overall_f1': 0.5182058047493403, 'overall_accuracy': 0.7024098848433915}
			------------EPOCH 27---------------
Loss:  tensor(4.0678, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.5271, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.3423, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.4919, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.0740, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.4307, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.6083, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.7948, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.2463, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.6355, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.0883, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.5843, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.6743, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.7115, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.2071, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.6267, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.4117, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.4784, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.5559, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.2462, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.8872, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.4917, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.4029, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.0137, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.6414, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.7483, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.3318, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.3395, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.6593, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.6276, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.8271, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.0190, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.4469, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.6219, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.7531, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.6233, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4511904761904762, 'recall': 0.4846547314578005, 'f1': 0.467324290998767, 'number': 782}, 'P': {'precision': 0.49105691056910566, 'recall': 0.6398305084745762, 'f1': 0.5556577736890524, 'number': 944}, 'overall_precision': 0.4748792270531401, 'overall_recall': 0.5695249130938587, 'overall_f1': 0.517913593256059, 'overall_accuracy': 0.7025915915233834}
			------------EPOCH 28---------------
Loss:  tensor(3.6208, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.9372, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.6383, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.8558, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.8020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.0531, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.2334, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.2973, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.9638, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.0806, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.2917, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.9132, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.5107, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.9976, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.8801, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.8849, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.0904, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.9517, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.2722, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.9060, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.6040, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.1296, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.6076, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.5350, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.3709, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.4574, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.9220, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.6650, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.4121, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.2712, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.6687, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.2799, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.8696, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.0861, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.3429, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.2825, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.45023696682464454, 'recall': 0.4859335038363171, 'f1': 0.4674046740467404, 'number': 782}, 'P': {'precision': 0.49228269699431354, 'recall': 0.6419491525423728, 'f1': 0.5572413793103448, 'number': 944}, 'overall_precision': 0.47518072289156627, 'overall_recall': 0.5712630359212051, 'overall_f1': 0.5188108392528282, 'overall_accuracy': 0.7032275649033548}
			------------EPOCH 29---------------
Loss:  tensor(3.2434, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.4669, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.9578, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.3397, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.5894, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.7956, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.9093, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.8469, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.4269, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.7037, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.7387, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.5861, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.4891, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.3985, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.6289, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.2370, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.8484, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.8739, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.1830, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.4685, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.3679, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.8043, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.8532, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.1574, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.1433, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.5749, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.3856, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.2152, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.9289, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.5546, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.6492, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.3191, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.3071, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.9809, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.9848, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.44799054373522457, 'recall': 0.4846547314578005, 'f1': 0.4656019656019656, 'number': 782}, 'P': {'precision': 0.4878444084278768, 'recall': 0.6377118644067796, 'f1': 0.5528007346189164, 'number': 944}, 'overall_precision': 0.4716346153846154, 'overall_recall': 0.5683661645422943, 'overall_f1': 0.515501839201261, 'overall_accuracy': 0.7027960115383742}
			------------EPOCH 30---------------
Loss:  tensor(2.9325, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.0497, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.2807, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.9493, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.4172, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.4602, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.6297, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.5233, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.8494, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.2603, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.3183, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.1191, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.5418, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.8750, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.3645, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.6829, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6659, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.3106, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.3828, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.0684, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.1722, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.5370, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.1861, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.8681, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.9336, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.9812, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.2986, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.3384, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.0386, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.5825, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.4603, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.1035, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.8661, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.8220, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.6825, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.7230, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4490521327014218, 'recall': 0.4846547314578005, 'f1': 0.4661746617466175, 'number': 782}, 'P': {'precision': 0.48906882591093115, 'recall': 0.6398305084745762, 'f1': 0.554382744378155, 'number': 944}, 'overall_precision': 0.4728234728234728, 'overall_recall': 0.5695249130938587, 'overall_f1': 0.516688567674113, 'overall_accuracy': 0.7027278715333772}


		-------------RUN 3-----------
			------------EPOCH 1---------------
Loss:  tensor(5926.6060, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2403.7534, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2577.6526, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1023.3906, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1661.1963, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2770.3125, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1739.5918, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3092.8831, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3053.8320, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3086.7090, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3396.6956, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2918.6367, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1255.5764, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2327.4302, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1453.4108, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1215.8792, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1495.7839, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2036.9128, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1459.5471, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2516.5000, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3197.6555, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2706.3833, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2257.1094, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3381.6687, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2400.2954, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2802.2017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1987.9199, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1817.8291, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1212.7439, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2139.0081, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1819.4261, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1532.7649, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1818.9055, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(836.6591, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2185.8145, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1157.7607, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(859.6386, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.025584472871636524, 'recall': 0.0849194729136164, 'f1': 0.03932203389830508, 'number': 683}, 'P': {'precision': 0.026490066225165563, 'recall': 0.004223864836325237, 'f1': 0.007285974499089253, 'number': 947}, 'overall_precision': 0.02564102564102564, 'overall_recall': 0.03803680981595092, 'overall_f1': 0.03063241106719368, 'overall_accuracy': 0.40427274080690784}
			------------EPOCH 2---------------
Loss:  tensor(4207.3613, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1609.3081, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1733.5020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(722.8083, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1310.1437, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2135.7612, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1310.5746, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2359.5901, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2406.4119, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2485.0288, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2668.8657, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2224.9771, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(993.8304, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1980.2679, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1171.8579, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(995.1249, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1313.3302, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1762.6121, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1259.8442, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2154.9971, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2800.2324, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2425.8545, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1934.2496, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2988.8174, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2244.1467, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2550.8547, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1793.1921, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1651.8759, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1096.1643, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2029.6150, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1534.1198, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1254.5486, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1586.9932, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(757.7595, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1819.5007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1049.0759, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(767.9764, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.04631416441528367, 'recall': 0.17569546120058566, 'f1': 0.07330482590103848, 'number': 683}, 'P': {'precision': 0.21122994652406418, 'recall': 0.16684266103484688, 'f1': 0.18643067846607672, 'number': 947}, 'overall_precision': 0.08325846061695118, 'overall_recall': 0.1705521472392638, 'overall_f1': 0.11189374119541155, 'overall_accuracy': 0.5351843580963724}
			------------EPOCH 3---------------
Loss:  tensor(3511.4517, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1358.2915, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1447.9585, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(586.4255, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1066.1667, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1761.4092, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1038.3668, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1842.9187, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2078.4285, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2213.5508, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2263.7014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1969.3818, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(814.5171, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1703.5269, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1020.4995, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(881.5753, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1099.9417, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1493.5613, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1143.7778, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1883.8219, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2331.9937, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1978.2692, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1727.1318, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2557.5947, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1896.2329, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2100.1616, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1455.3083, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1376.3975, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(927.1552, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1668.1924, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1204.8438, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(951.5347, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1312.6498, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(629.3197, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1540.8491, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(878.9420, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(627.6712, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.07336417713152676, 'recall': 0.16251830161054173, 'f1': 0.10109289617486339, 'number': 683}, 'P': {'precision': 0.34701492537313433, 'recall': 0.3928194297782471, 'f1': 0.36849925705794945, 'number': 947}, 'overall_precision': 0.18684719535783365, 'overall_recall': 0.296319018404908, 'overall_f1': 0.22918149466192173, 'overall_accuracy': 0.6241129472482755}
			------------EPOCH 4---------------
Loss:  tensor(3020.6311, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1166.4691, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1201.1335, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(511.8735, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(914.7355, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1505.0579, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(807.4441, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1431.9052, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1703.8289, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1903.3120, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1847.1289, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1654.2268, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(621.4784, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1511.2104, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(882.2451, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(720.7701, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(947.6175, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1263.8630, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(982.0336, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1605.7749, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1954.5339, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1519.9976, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1463.8054, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2059.8193, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1596.9329, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1542.3621, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1030.0326, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1027.2910, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(708.5901, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1252.3022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(838.7053, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(628.7705, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1087.5952, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(493.8001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1259.4568, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(715.3259, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(478.5500, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.15032154340836013, 'recall': 0.27379209370424595, 'f1': 0.19408406850025947, 'number': 683}, 'P': {'precision': 0.4116504854368932, 'recall': 0.4477296726504752, 'f1': 0.42893272635306023, 'number': 947}, 'overall_precision': 0.26868953386103783, 'overall_recall': 0.37484662576687117, 'overall_f1': 0.3130122950819672, 'overall_accuracy': 0.6752022232147288}
			------------EPOCH 5---------------
Loss:  tensor(2447.2717, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(976.4781, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(927.0384, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(422.2886, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(706.7416, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1150.5415, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(573.9598, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1048.7416, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1368.3839, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1543.3801, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1391.3083, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1319.8035, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(412.1605, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1262.0544, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(720.5487, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(499.3660, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(790.4590, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(880.7344, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(736.0939, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1244.5215, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1492.7913, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(772.4104, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1079.2356, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1510.6442, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1219.2476, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1108.7612, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(646.3089, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(550.0206, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(473.5522, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(712.1707, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(595.3221, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(429.3716, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(874.3619, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(364.6591, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(915.9819, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(587.2424, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(316.5898, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.1884272997032641, 'recall': 0.18594436310395315, 'f1': 0.18717759764185707, 'number': 683}, 'P': {'precision': 0.40400925212027755, 'recall': 0.5533262935586061, 'f1': 0.4670231729055258, 'number': 947}, 'overall_precision': 0.3302891933028919, 'overall_recall': 0.3993865030674847, 'overall_f1': 0.3615662316023327, 'overall_accuracy': 0.6914793310505682}
			------------EPOCH 6---------------
Loss:  tensor(1901.2925, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(759.7672, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(567.1027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(331.1723, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(551.8739, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(806.6565, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(436.3759, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(828.9436, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(899.8859, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1132.7307, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1013.8069, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(857.5673, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(273.7926, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1007.9084, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(522.0981, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(388.0686, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(538.1995, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(694.8562, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(476.2266, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(897.5721, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1029.6185, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(394.2128, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(746.8394, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1000.1945, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(900.1287, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(750.4189, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(424.2662, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(312.5270, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(345.0170, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(488.8365, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(427.6736, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(395.0156, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(649.6136, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(265.2332, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(673.5995, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(521.0030, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(217.6199, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.23153526970954358, 'recall': 0.4084919472913616, 'f1': 0.29555084745762716, 'number': 683}, 'P': {'precision': 0.40160642570281124, 'recall': 0.42238648363252373, 'f1': 0.41173443129181675, 'number': 947}, 'overall_precision': 0.3084961381190368, 'overall_recall': 0.4165644171779141, 'overall_f1': 0.3544766379535369, 'overall_accuracy': 0.6607116272145303}
			------------EPOCH 7---------------
Loss:  tensor(1341.9099, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(502.4111, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(434.1129, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(237.9278, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(453.4924, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(685.3857, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(412.5547, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(572.8544, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(690.8143, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(803.1791, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(738.8088, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(654.0234, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(209.0280, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(675.0503, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(381.8331, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(237.1717, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(344.1112, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(493.7034, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(294.9296, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(658.1288, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1050.2673, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(374.3181, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1026.8922, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(943.7406, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(552.4370, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(486.4279, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(264.1736, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(238.3052, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(257.1176, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(398.3594, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(300.7703, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(268.9854, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(708.6089, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(364.4884, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(785.8220, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(651.9929, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(187.5482, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.26294820717131473, 'recall': 0.28989751098096633, 'f1': 0.2757660167130919, 'number': 683}, 'P': {'precision': 0.46285289747399705, 'recall': 0.6578669482576558, 'f1': 0.5433929350196249, 'number': 947}, 'overall_precision': 0.39113863744640304, 'overall_recall': 0.503680981595092, 'overall_f1': 0.4403325288281041, 'overall_accuracy': 0.7120490298248225}
			------------EPOCH 8---------------
Loss:  tensor(1062.5614, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(362.3123, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(237.3873, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(175.2172, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(306.0712, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(441.6058, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(225.0330, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(321.2468, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(420.3193, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(555.2378, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(556.6368, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(409.5565, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(123.9733, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(607.1866, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(279.2659, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(206.8409, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(248.6467, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(458.4706, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(227.6085, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(538.3783, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(602.5924, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(217.7683, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(482.8265, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(499.5125, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(513.5948, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(489.1742, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(372.8542, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(261.3918, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(145.3071, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(190.5269, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(341.4594, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(115.1947, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(413.0700, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(129.8840, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(375.1199, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(238.0339, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(215.3501, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.14928909952606634, 'recall': 0.09224011713030747, 'f1': 0.11402714932126697, 'number': 683}, 'P': {'precision': 0.4612141652613828, 'recall': 0.5776135163674763, 'f1': 0.512892639474918, 'number': 947}, 'overall_precision': 0.3793532338308458, 'overall_recall': 0.37423312883435583, 'overall_f1': 0.37677578752316243, 'overall_accuracy': 0.6599672472830133}
			------------EPOCH 9---------------
Loss:  tensor(2803.4731, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(789.9717, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(805.6216, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(176.9091, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(266.6292, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(302.8422, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(237.2602, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(321.4243, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(390.5690, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(524.9949, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(416.4389, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(482.5059, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(166.2588, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(675.1780, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(363.7499, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(176.9667, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(406.4818, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(807.8276, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(263.4051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(492.3250, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1328.3923, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1306.8102, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(877.6194, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1122.6960, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1088.9595, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1613.3550, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(716.8893, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(468.1001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(124.1831, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(353.9604, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(258.8297, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(136.8434, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(492.9982, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(187.5162, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(464.8170, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(281.1759, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(166.0628, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.1695464362850972, 'recall': 0.22986822840409957, 'f1': 0.19515226848974518, 'number': 683}, 'P': {'precision': 0.4704225352112676, 'recall': 0.529039070749736, 'f1': 0.4980119284294235, 'number': 947}, 'overall_precision': 0.3304871923656454, 'overall_recall': 0.403680981595092, 'overall_f1': 0.3634355150510909, 'overall_accuracy': 0.6846310356806113}
			------------EPOCH 10---------------
Loss:  tensor(845.0583, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(353.2232, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(238.4076, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(186.2667, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(409.7052, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(507.8005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(358.8433, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(598.0220, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(815.5359, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1369.4131, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1072.0901, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1159.9984, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(443.4367, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1042.7600, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(545.6401, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(428.8586, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(661.3899, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(823.6831, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(610.5901, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1123.9534, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(902.6327, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(628.3342, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(994.9569, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1073.7136, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(551.6278, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(305.0265, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(186.8427, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(230.5658, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(118.6951, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(258.2669, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(151.6412, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(112.4281, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(724.5533, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(295.1973, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(635.7716, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(471.3733, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(427.1277, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.220703125, 'recall': 0.6617862371888726, 'f1': 0.3310142804833394, 'number': 683}, 'P': {'precision': 0.5502958579881657, 'recall': 0.09820485744456177, 'f1': 0.16666666666666669, 'number': 947}, 'overall_precision': 0.2458276950834461, 'overall_recall': 0.3343558282208589, 'overall_f1': 0.283337665713543, 'overall_accuracy': 0.48027393181479827}
			------------EPOCH 11---------------
Loss:  tensor(1793.5615, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1009.0249, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(711.6240, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(349.3643, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(695.1398, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1058.8979, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(432.1222, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(518.6672, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(652.6058, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(803.2106, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(476.8170, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(703.8958, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(102.8088, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(413.7503, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(253.5265, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(122.6530, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(282.9314, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(336.3640, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(187.8018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(353.2879, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(584.7206, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(297.1248, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(526.7316, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(734.8154, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(696.0402, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(567.7495, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(406.0486, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(446.9222, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(302.1667, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(370.8227, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(330.0128, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(215.3445, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(340.7373, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(117.8685, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(367.5746, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(252.4178, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(128.3480, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.2566943674976916, 'recall': 0.40702781844802344, 'f1': 0.31483578708946774, 'number': 683}, 'P': {'precision': 0.5201900237529691, 'recall': 0.4625131995776135, 'f1': 0.4896590273896031, 'number': 947}, 'overall_precision': 0.37194805194805197, 'overall_recall': 0.4392638036809816, 'overall_f1': 0.40281293952180036, 'overall_accuracy': 0.6888243759614907}
			------------EPOCH 12---------------
Loss:  tensor(715.1204, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(374.4682, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(226.3543, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(135.8398, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(310.3757, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(350.8517, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(199.8479, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(305.8597, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(619.8197, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(669.3809, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(752.9464, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1005.6586, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(227.4271, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(698.6454, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(396.4872, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(206.5137, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(205.0505, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(453.9317, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(227.3914, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(445.1653, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(331.3783, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(155.5988, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(306.9442, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(328.2106, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(283.4117, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(249.3902, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(145.4044, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(157.3130, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(103.0111, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(149.1433, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(86.8533, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(91.5285, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(372.2256, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(162.1307, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(305.6897, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(217.0437, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(149.8416, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.2828870779976717, 'recall': 0.3557833089311859, 'f1': 0.3151750972762646, 'number': 683}, 'P': {'precision': 0.38839590443686006, 'recall': 0.600844772967265, 'f1': 0.47180762852404634, 'number': 947}, 'overall_precision': 0.3493975903614458, 'overall_recall': 0.498159509202454, 'overall_f1': 0.4107233181588265, 'overall_accuracy': 0.6854002282765124}
			------------EPOCH 13---------------
Loss:  tensor(1127.9340, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(387.1191, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(479.0267, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(143.3349, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(440.9658, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(492.1991, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(310.6381, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(454.4579, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(571.5477, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(643.6088, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(454.3670, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(533.8268, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(90.0586, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(359.2220, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(181.5414, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(88.7410, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(184.7163, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(259.2942, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(145.2750, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(268.3561, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(343.1512, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(207.6772, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(311.0949, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(304.0634, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(282.4528, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(271.9831, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(164.1676, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(121.6008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(72.4636, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(107.9401, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(79.5416, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(66.8126, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(245.5018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(104.5313, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(223.5154, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(170.9480, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(107.7317, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.31754256106587714, 'recall': 0.6281112737920937, 'f1': 0.42182890855457233, 'number': 683}, 'P': {'precision': 0.568503937007874, 'recall': 0.3812038014783527, 'f1': 0.4563843236409609, 'number': 947}, 'overall_precision': 0.39778449144008055, 'overall_recall': 0.48466257668711654, 'overall_f1': 0.43694690265486724, 'overall_accuracy': 0.6627710783583941}
			------------EPOCH 14---------------
Loss:  tensor(374.5285, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(171.2427, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(124.1728, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(117.7365, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(157.9303, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(164.2129, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(100.4418, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(121.1700, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(191.5905, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(232.4335, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(309.9808, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(254.1747, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(75.0738, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(241.2379, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(177.9803, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(82.5891, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(135.6988, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(238.0735, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(107.5563, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(204.0098, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(243.6124, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(107.1373, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(144.0700, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(222.5757, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(235.8972, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(165.7660, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(128.8674, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(121.4002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(66.0241, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(96.0151, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.3006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.3082, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(142.0609, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.8599, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(113.3610, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(79.1903, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.9925, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.42126514131897713, 'recall': 0.4582723279648609, 'f1': 0.43899018232819076, 'number': 683}, 'P': {'precision': 0.47590870667793744, 'recall': 0.5945089757127772, 'f1': 0.5286384976525821, 'number': 947}, 'overall_precision': 0.45482866043613707, 'overall_recall': 0.5374233128834356, 'overall_f1': 0.49268841394825647, 'overall_accuracy': 0.7395662746265694}
			------------EPOCH 15---------------
Loss:  tensor(284.9527, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(79.3289, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(88.7565, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.8692, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(96.7277, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(115.0789, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(94.1471, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(81.6823, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(184.7381, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(201.1375, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(254.5156, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(207.0946, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.6025, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(194.0590, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(112.4850, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.9756, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(98.7633, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(123.2904, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(65.7529, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(118.2332, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(152.6986, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(61.9800, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(88.6494, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(130.9245, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(128.7990, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(111.0283, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(76.8048, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(95.0057, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.6284, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(69.7910, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.4914, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.6540, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(102.7362, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.7128, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(84.9018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.0879, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.2690, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.35188679245283017, 'recall': 0.5461200585651538, 'f1': 0.4279977051061388, 'number': 683}, 'P': {'precision': 0.4990059642147117, 'recall': 0.5300950369588173, 'f1': 0.5140809011776752, 'number': 947}, 'overall_precision': 0.4235237173281704, 'overall_recall': 0.5368098159509203, 'overall_f1': 0.4734848484848485, 'overall_accuracy': 0.7076075629001042}
			------------EPOCH 16---------------
Loss:  tensor(258.8668, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.7136, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(81.1788, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.2485, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(73.3632, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(92.1636, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.1061, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.3311, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(115.3636, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(100.1542, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(197.0669, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(118.3972, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.5968, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(109.5463, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(77.1328, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.0809, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(72.4145, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(82.1467, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.9597, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(97.9376, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(124.6497, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(59.8562, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(79.4587, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(109.6517, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(95.2391, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(82.9804, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(68.2948, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(72.8088, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.4763, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(61.6112, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.1569, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.0730, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(79.4595, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.3412, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(65.2412, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.4368, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.9704, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.373320537428023, 'recall': 0.5695461200585652, 'f1': 0.45101449275362315, 'number': 683}, 'P': {'precision': 0.5317545748116254, 'recall': 0.5216473072861668, 'f1': 0.5266524520255863, 'number': 947}, 'overall_precision': 0.44799594114662605, 'overall_recall': 0.5417177914110429, 'overall_f1': 0.49041932796445425, 'overall_accuracy': 0.7105106446330207}
			------------EPOCH 17---------------
Loss:  tensor(206.2119, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.2999, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(67.4471, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.0694, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.4572, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(70.6587, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.9637, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.7077, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(91.1817, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(79.6868, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(165.9901, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(88.2202, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.1761, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(82.3550, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(62.3724, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.0942, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.7896, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(62.4354, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.2820, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(65.7687, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(99.4384, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.3038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(59.0086, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(92.0743, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(70.4981, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.8208, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.8504, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.3309, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.4297, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.0033, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.8631, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.7543, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.1152, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.4696, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.0937, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.6784, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.0913, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3879310344827586, 'recall': 0.527086383601757, 'f1': 0.44692737430167595, 'number': 683}, 'P': {'precision': 0.5354645354645354, 'recall': 0.5659978880675819, 'f1': 0.5503080082135523, 'number': 947}, 'overall_precision': 0.4644893727319855, 'overall_recall': 0.5496932515337424, 'overall_f1': 0.5035122225344197, 'overall_accuracy': 0.720783087687956}
			------------EPOCH 18---------------
Loss:  tensor(161.2487, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.3221, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.6234, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.1392, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.5461, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(59.8537, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.7212, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.9244, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(74.5366, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(71.1844, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(152.8811, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(71.4098, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.9751, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(64.2020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.2742, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.6255, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.9864, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.1668, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.6983, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.7683, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(83.0753, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.1180, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.3985, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(85.9296, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.6391, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.6833, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.8110, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.6689, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.4724, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.0170, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.3647, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.7715, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.6052, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.2384, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.7575, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.2838, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.4572, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3847826086956522, 'recall': 0.5183016105417276, 'f1': 0.4416718652526513, 'number': 683}, 'P': {'precision': 0.5339147286821705, 'recall': 0.5818373812038015, 'f1': 0.5568468923698837, 'number': 947}, 'overall_precision': 0.4636270491803279, 'overall_recall': 0.5552147239263804, 'overall_f1': 0.5053042992741484, 'overall_accuracy': 0.7261922485236465}
			------------EPOCH 19---------------
Loss:  tensor(130.7020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.1490, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.0725, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.7050, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.7483, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.8880, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.5438, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.1921, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.5361, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(64.4490, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(144.7663, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(59.4821, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.6525, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.0266, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.5003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.0774, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.4514, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.5858, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.1595, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.3384, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(71.9851, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.2027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.1054, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(81.3081, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.3893, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.7971, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.7550, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.1037, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.6259, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.7826, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.5838, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.7235, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.7109, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.4021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.8186, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.5823, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.6289, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.384204909284952, 'recall': 0.527086383601757, 'f1': 0.4444444444444445, 'number': 683}, 'P': {'precision': 0.555667001003009, 'recall': 0.5850052798310454, 'f1': 0.5699588477366254, 'number': 947}, 'overall_precision': 0.4725956566701138, 'overall_recall': 0.5607361963190184, 'overall_f1': 0.5129068462401796, 'overall_accuracy': 0.7243312986948538}
			------------EPOCH 20---------------
Loss:  tensor(107.8956, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.2159, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.3881, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.2512, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.0058, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.1583, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.0953, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.9239, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.6746, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.6681, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(130.7681, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.9957, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.4094, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.2580, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.9601, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.3136, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.9787, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.9117, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.3774, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.5426, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(66.9939, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.6789, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.0421, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(73.8148, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.0510, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.4490, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.3346, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.4622, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.4866, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.7383, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.2656, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.7933, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.9793, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.5322, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.0759, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.5174, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.2320, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3887113951011715, 'recall': 0.5344070278184481, 'f1': 0.45006165228113443, 'number': 683}, 'P': {'precision': 0.5608519269776876, 'recall': 0.5839493136219641, 'f1': 0.5721676151060527, 'number': 947}, 'overall_precision': 0.4768831168831169, 'overall_recall': 0.5631901840490797, 'overall_f1': 0.5164556962025316, 'overall_accuracy': 0.7229914148181231}
			------------EPOCH 21---------------
Loss:  tensor(89.7830, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.6917, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.2266, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.8424, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.2612, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.0961, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.6975, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.6448, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.6196, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.9455, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(109.6488, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.7874, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.1586, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.9309, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.4146, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.1677, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.3012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.1136, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.2529, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.6884, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(59.8599, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.6196, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.6447, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(68.2807, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.7561, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.2144, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.9164, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.2953, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.6625, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.6274, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.4127, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.9037, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.7911, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.5239, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.0258, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.6138, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.6843, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.39544962080173346, 'recall': 0.5344070278184481, 'f1': 0.45454545454545453, 'number': 683}, 'P': {'precision': 0.5475475475475475, 'recall': 0.5776135163674763, 'f1': 0.5621788283658787, 'number': 947}, 'overall_precision': 0.4745057232049948, 'overall_recall': 0.5595092024539877, 'overall_f1': 0.5135135135135135, 'overall_accuracy': 0.7243312986948538}
			------------EPOCH 22---------------
Loss:  tensor(76.8341, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.3154, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.1143, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.3785, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.6975, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.9249, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.3114, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.5328, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.1829, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.9488, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(85.0667, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.4456, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.3577, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.8471, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.9421, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.8163, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.2018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.0290, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.9726, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.7658, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.4191, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.1541, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.5543, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(65.0380, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.7100, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.9916, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.9576, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.4667, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.6526, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.9322, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.8971, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.0875, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.6692, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.4846, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.5679, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.4210, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.6888, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3887113951011715, 'recall': 0.5344070278184481, 'f1': 0.45006165228113443, 'number': 683}, 'P': {'precision': 0.5551061678463094, 'recall': 0.5797254487856388, 'f1': 0.5671487603305784, 'number': 947}, 'overall_precision': 0.47406639004149376, 'overall_recall': 0.5607361963190184, 'overall_f1': 0.5137717818999438, 'overall_accuracy': 0.7234132301126495}
			------------EPOCH 23---------------
Loss:  tensor(67.1873, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.3245, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.4628, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.8878, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.7662, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.3258, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.6589, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.2657, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.5151, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.7703, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(72.1048, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.7353, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.3797, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.9214, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.2713, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.8428, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.6383, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.4954, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.3835, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.9827, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.8819, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.9235, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.8669, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(72.9678, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.6429, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.4720, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.6348, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.9472, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.5418, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.9987, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.2948, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.7681, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.3128, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.4506, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.6001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.4786, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.9152, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.39013933547695606, 'recall': 0.5329428989751098, 'f1': 0.45049504950495045, 'number': 683}, 'P': {'precision': 0.5584677419354839, 'recall': 0.5850052798310454, 'f1': 0.5714285714285715, 'number': 947}, 'overall_precision': 0.4768831168831169, 'overall_recall': 0.5631901840490797, 'overall_f1': 0.5164556962025316, 'overall_accuracy': 0.7229666021537393}
			------------EPOCH 24---------------
Loss:  tensor(60.7804, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.9189, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.1641, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.5381, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.1766, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.6358, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.9554, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.4981, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.1037, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.3058, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.6853, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.8837, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.4993, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.6384, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.6621, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.6647, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.6739, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.0441, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.3731, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.5734, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.5934, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.7561, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.8885, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(69.2163, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.4951, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.0511, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.2459, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.4696, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.2517, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.6261, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.2020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.2402, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.8423, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.4640, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.2488, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.3974, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.8418, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.39565217391304347, 'recall': 0.5329428989751098, 'f1': 0.45414847161572053, 'number': 683}, 'P': {'precision': 0.5677290836653387, 'recall': 0.6019007391763463, 'f1': 0.5843157355202461, 'number': 947}, 'overall_precision': 0.48544698544698545, 'overall_recall': 0.5730061349693252, 'overall_f1': 0.5256049521665729, 'overall_accuracy': 0.7250756786263709}
			------------EPOCH 25---------------
Loss:  tensor(55.7679, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.2294, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.0808, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.5494, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.8502, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.3455, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.9583, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.0781, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.5932, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.8125, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.5863, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.5598, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.4467, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.2436, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.3732, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.4399, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.4964, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.6558, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.4218, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.4344, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.8284, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.3119, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.2287, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(68.0937, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.6417, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.8581, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.1208, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.4082, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.1236, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.1193, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.1488, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.0683, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.7876, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.9719, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.2833, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.3862, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.4067, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3626373626373626, 'recall': 0.48316251830161056, 'f1': 0.4143126177024482, 'number': 683}, 'P': {'precision': 0.5298726738491675, 'recall': 0.5712777191129884, 'f1': 0.5497967479674797, 'number': 947}, 'overall_precision': 0.4510616261004661, 'overall_recall': 0.5343558282208589, 'overall_f1': 0.4891884302162314, 'overall_accuracy': 0.7202868344002779}
			------------EPOCH 26---------------
Loss:  tensor(137.9429, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.8305, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(64.0956, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.9834, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.5065, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.8181, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.5425, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.7542, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.7691, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.9968, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(119.7817, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(65.9717, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.7798, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(82.5643, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.4333, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.9138, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.6231, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(59.8765, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.7089, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(128.9144, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(137.2453, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.6303, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.4343, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(85.0354, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.6255, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.3447, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.5163, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.7157, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.6604, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.2426, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.4294, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.5682, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.6260, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.1236, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.3810, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.2587, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.9660, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.36587771203155817, 'recall': 0.5431918008784773, 'f1': 0.43724219210371246, 'number': 683}, 'P': {'precision': 0.5374753451676528, 'recall': 0.5755015839493136, 'f1': 0.5558388577256501, 'number': 947}, 'overall_precision': 0.4516765285996055, 'overall_recall': 0.561963190184049, 'overall_f1': 0.5008201202843083, 'overall_accuracy': 0.7169123120440672}
			------------EPOCH 27---------------
Loss:  tensor(115.1239, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.7333, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.3813, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.8080, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.3203, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.9620, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.9126, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.0088, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.6653, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.1177, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(78.0221, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.3448, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.3354, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.8097, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.6797, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.2373, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.0745, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.2403, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.3264, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.6742, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.5707, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.1600, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.9441, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.4662, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.3756, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.1450, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.5861, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.9095, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.5219, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.6724, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.5421, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.9530, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.2882, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.1178, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.4815, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.3033, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.1717, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4019933554817276, 'recall': 0.5314787701317716, 'f1': 0.45775535939470363, 'number': 683}, 'P': {'precision': 0.5537848605577689, 'recall': 0.587117212249208, 'f1': 0.5699641209636083, 'number': 947}, 'overall_precision': 0.4819087572102779, 'overall_recall': 0.5638036809815951, 'overall_f1': 0.5196494204127792, 'overall_accuracy': 0.7288720162771078}
			------------EPOCH 28---------------
Loss:  tensor(89.4545, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.9489, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.9373, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.3123, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.4433, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.0445, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.8146, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.2481, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.3917, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.4620, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.5652, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.8792, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.3839, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.7355, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.3694, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.1078, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.9389, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.5130, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.5751, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.6421, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.3484, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.1095, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.5595, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.1629, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.0974, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.6108, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.8430, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.1450, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.9672, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.4375, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.1022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.9862, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.0613, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.2517, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.7710, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.0960, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.7990, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4067245119305857, 'recall': 0.5490483162518301, 'f1': 0.4672897196261682, 'number': 683}, 'P': {'precision': 0.5730337078651685, 'recall': 0.5923970432946146, 'f1': 0.5825545171339565, 'number': 947}, 'overall_precision': 0.49237243556023147, 'overall_recall': 0.5742331288343558, 'overall_f1': 0.5301614273576891, 'overall_accuracy': 0.7267877524688601}
			------------EPOCH 29---------------
Loss:  tensor(55.2468, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.5099, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.6723, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.7357, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.0232, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.9150, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.2210, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.0535, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.4033, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.2505, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.9569, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.7420, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.2319, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.0707, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.9072, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.2908, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.7518, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.7939, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.1601, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.4528, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.8532, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.6156, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.3568, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.4621, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.5386, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.9267, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.8691, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.4821, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.8005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.7614, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.0867, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.8961, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.1470, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.6974, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.1523, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.1822, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.0745, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4110032362459547, 'recall': 0.5578330893118595, 'f1': 0.47329192546583854, 'number': 683}, 'P': {'precision': 0.5689655172413793, 'recall': 0.5923970432946146, 'f1': 0.5804449042938438, 'number': 947}, 'overall_precision': 0.4924202822791427, 'overall_recall': 0.5779141104294478, 'overall_f1': 0.5317527519051651, 'overall_accuracy': 0.7274576944072254}
			------------EPOCH 30---------------
Loss:  tensor(45.3841, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.6282, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.1083, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.1390, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.8476, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.4814, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.4505, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.7799, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.0320, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.9858, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.9563, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.9671, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.1884, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.6582, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.4934, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.6820, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.1398, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.9049, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.5354, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.0952, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.1869, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.7714, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.2593, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.9735, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.3014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.7065, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.9612, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.9139, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.5933, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.3423, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.3951, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.2548, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.2321, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.9916, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.7984, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.0962, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.7627, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4022222222222222, 'recall': 0.5300146412884333, 'f1': 0.45735944409349333, 'number': 683}, 'P': {'precision': 0.5528375733855186, 'recall': 0.5966209081309398, 'f1': 0.5738953783646522, 'number': 947}, 'overall_precision': 0.48231009365244537, 'overall_recall': 0.5687116564417178, 'overall_f1': 0.5219594594594595, 'overall_accuracy': 0.7305096521264454}


		-------------RUN 4-----------
			------------EPOCH 1---------------
Loss:  tensor(2068.2920, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1193.5063, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1651.3542, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1427.6809, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3416.8862, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3137.8320, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2059.1023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3198.1040, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2512.1907, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1782.0586, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1890.2285, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1795.1638, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2504.4619, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2816.7461, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1436.4786, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2858.0256, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3135.5117, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2654.5435, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2667.2859, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2353.1226, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1733.8474, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(501.1111, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1158.1416, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1998.8041, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1571.4338, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2049.1387, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2504.8552, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2022.9263, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2702.6860, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1486.1909, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2153.6047, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2785.8699, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1019.8512, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(965.9638, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'number': 854}, 'P': {'precision': 0.03139013452914798, 'recall': 0.008403361344537815, 'f1': 0.013257575757575756, 'number': 833}, 'overall_precision': 0.02456140350877193, 'overall_recall': 0.004149377593360996, 'overall_f1': 0.007099391480730223, 'overall_accuracy': 0.3318533404996144}
			------------EPOCH 2---------------
Loss:  tensor(1325.2012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(793.3037, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1050.8665, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(830.2172, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2094.7896, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2147.0330, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1477.2532, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2378.7322, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2042.5769, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1292.8508, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1336.3679, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1398.8369, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2078.0239, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2399.8682, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1223.0996, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2408.5974, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2821.0112, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2260.4097, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2187.7385, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2032.7231, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1493.0220, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(419.9264, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1078.8428, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1877.0801, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1361.9564, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1809.2046, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2083.6719, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1725.4744, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2381.9707, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1252.2500, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1893.1548, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2457.9023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(874.6344, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(766.5362, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.1477832512315271, 'recall': 0.0702576112412178, 'f1': 0.09523809523809525, 'number': 854}, 'P': {'precision': 0.10496338486574451, 'recall': 0.15486194477791115, 'f1': 0.12512124151309406, 'number': 833}, 'overall_precision': 0.11559633027522936, 'overall_recall': 0.11203319502074689, 'overall_f1': 0.11378687537627935, 'overall_accuracy': 0.4953146542658846}
			------------EPOCH 3---------------
Loss:  tensor(1140.5847, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(692.4341, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(948.1431, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(679.6797, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1713.3247, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1893.3441, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1268.8613, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2135.7112, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1862.0588, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1056.6412, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1140.1768, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1150.9388, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1842.0221, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2103.2014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1038.9200, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2046.2256, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2397.4746, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1750.7135, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1768.5289, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1749.1218, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1313.5598, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(360.0241, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1000.0626, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1640.1780, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1103.7292, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1411.1936, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1755.1108, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1468.8296, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1836.2417, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1021.3656, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1536.6721, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2104.2625, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(769.2428, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(676.3203, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.13043478260869565, 'recall': 0.09133489461358314, 'f1': 0.10743801652892561, 'number': 854}, 'P': {'precision': 0.11550976138828634, 'recall': 0.25570228091236497, 'f1': 0.15913335823683228, 'number': 833}, 'overall_precision': 0.11916461916461916, 'overall_recall': 0.17249555423829283, 'overall_f1': 0.14095422620489223, 'overall_accuracy': 0.54730913934522}
			------------EPOCH 4---------------
Loss:  tensor(922.7368, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(595.7930, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(832.7354, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(491.8764, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1158.6835, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1520.7783, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1034.4304, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1829.2356, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1609.1995, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(755.0938, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(885.0775, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(925.5485, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1571.9788, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1722.4679, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(816.4186, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1646.1296, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1932.1816, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1314.2695, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1387.6960, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1404.8253, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1100.9768, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(289.4588, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(832.8510, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1287.1846, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(937.3737, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1130.5918, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1550.6970, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1245.0227, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1466.3706, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(748.0051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1269.6912, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1667.6613, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(657.2018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(544.7921, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.20364741641337386, 'recall': 0.07845433255269321, 'f1': 0.11327134404057482, 'number': 854}, 'P': {'precision': 0.21895424836601307, 'recall': 0.4021608643457383, 'f1': 0.2835378755818874, 'number': 833}, 'overall_precision': 0.2162452931683701, 'overall_recall': 0.23829282750444578, 'overall_f1': 0.22673434856175975, 'overall_accuracy': 0.5467482999556003}
			------------EPOCH 5---------------
Loss:  tensor(708.5469, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(469.2443, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(658.5538, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(334.7463, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(840.9290, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1270.8679, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(753.7438, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1379.8494, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1326.4967, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(602.5701, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(669.2419, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(821.4584, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1288.7373, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1396.0580, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(631.8210, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1274.4688, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1581.5376, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1035.3132, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1099.1769, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1156.9901, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(875.4621, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(217.5682, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(642.0005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1011.0292, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(766.0966, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(914.6583, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1325.5820, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(979.9808, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1176.2273, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(560.1938, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1111.6976, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1443.5812, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(597.5172, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(448.7525, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.1791044776119403, 'recall': 0.09836065573770492, 'f1': 0.126984126984127, 'number': 854}, 'P': {'precision': 0.2622814321398834, 'recall': 0.37815126050420167, 'f1': 0.30973451327433627, 'number': 833}, 'overall_precision': 0.23892215568862277, 'overall_recall': 0.23651452282157676, 'overall_f1': 0.23771224307417335, 'overall_accuracy': 0.5535251092468394}
			------------EPOCH 6---------------
Loss:  tensor(555.6554, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(392.4737, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(524.8882, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(265.3901, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(736.8126, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1032.3074, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(527.9030, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1010.0690, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(888.7546, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(551.8915, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(497.2184, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(753.4934, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1289.4031, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1395.6960, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(607.5062, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1323.7834, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1654.5166, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1005.8069, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1121.0950, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1261.2012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(660.1910, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(175.5967, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(583.2098, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(949.9546, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(580.8164, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(725.6535, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1032.1046, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(733.5036, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1104.2397, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(467.1981, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1073.8635, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1393.1287, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(706.9044, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(507.6636, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.2070393374741201, 'recall': 0.351288056206089, 'f1': 0.26052974381241856, 'number': 854}, 'P': {'precision': 0.029411764705882353, 'recall': 0.004801920768307323, 'f1': 0.00825593395252838, 'number': 833}, 'overall_precision': 0.19179810725552052, 'overall_recall': 0.18020154119739182, 'overall_f1': 0.1858190709046455, 'overall_accuracy': 0.4783492627298857}
			------------EPOCH 7---------------
Loss:  tensor(824.6556, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(470.1138, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(747.5338, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(574.7964, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1486.7522, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1519.4117, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(763.8722, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1351.8339, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1142.8237, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(785.2042, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(769.2437, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(796.4833, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1298.4406, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1453.8245, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(727.4683, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1446.2429, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1462.2828, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1049.1492, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1033.5220, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1083.1481, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(837.1351, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(225.7695, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(650.7169, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1158.8134, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(957.3209, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1046.1505, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1734.9817, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1331.2128, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1333.7941, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(917.6381, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1375.9833, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2076.6206, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(792.4875, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(713.5230, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.2932330827067669, 'recall': 0.22833723653395785, 'f1': 0.2567478604344964, 'number': 854}, 'P': {'precision': 0.2501395868230039, 'recall': 0.5378151260504201, 'f1': 0.3414634146341463, 'number': 833}, 'overall_precision': 0.26180781758957655, 'overall_recall': 0.3811499703615886, 'overall_f1': 0.3104030895486362, 'overall_accuracy': 0.5834832799756969}
			------------EPOCH 8---------------
Loss:  tensor(692.5272, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(526.7224, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(600.7300, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(362.4812, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(767.5663, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1151.0354, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(849.7549, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1589.9426, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1147.4717, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(709.4133, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(749.3455, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(658.8151, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1116.3040, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1387.9834, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(530.1914, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1127.7139, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1262.1421, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(879.3143, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(872.6335, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(911.2518, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(653.1771, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(164.0287, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(484.5538, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(733.8687, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(626.1646, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(788.3115, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(948.4109, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(663.6050, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(771.3813, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(550.5057, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1118.0464, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1202.8121, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(510.4695, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(398.6592, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.18453865336658354, 'recall': 0.08665105386416862, 'f1': 0.11792828685258964, 'number': 854}, 'P': {'precision': 0.28387650085763294, 'recall': 0.39735894357743096, 'f1': 0.3311655827913957, 'number': 833}, 'overall_precision': 0.25845564773452456, 'overall_recall': 0.24007113218731477, 'overall_f1': 0.24892440073755376, 'overall_accuracy': 0.5686210361507723}
			------------EPOCH 9---------------
Loss:  tensor(350.1056, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(310.5372, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(309.5294, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(172.6679, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(490.3268, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(673.5711, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(277.9514, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(555.3351, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(446.1929, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(386.3958, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(344.2206, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(462.7315, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(624.5211, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(777.4905, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(387.6396, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(734.3823, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(883.5413, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(710.8779, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(749.0077, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(645.0255, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(445.0698, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(110.0455, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(403.5620, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(570.3508, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(360.2921, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(607.4722, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(669.1954, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(485.5766, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(352.3727, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(284.3930, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(868.1987, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(859.6367, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(335.3651, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(233.0476, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.336866902237927, 'recall': 0.33489461358313816, 'f1': 0.33587786259541985, 'number': 854}, 'P': {'precision': 0.34142752023546724, 'recall': 0.5570228091236494, 'f1': 0.4233576642335767, 'number': 833}, 'overall_precision': 0.33967391304347827, 'overall_recall': 0.44457617071724953, 'overall_f1': 0.3851091142490372, 'overall_accuracy': 0.6263874932816115}
			------------EPOCH 10---------------
Loss:  tensor(260.5923, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(219.1688, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(257.7008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(199.2840, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(346.5299, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(624.3330, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(241.8941, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(501.0264, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(338.8882, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(285.9122, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(271.5957, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(298.4221, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(403.3984, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(521.6323, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(249.2220, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(536.0031, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(683.3855, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(510.8255, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(517.8157, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(445.4552, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(359.6038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(76.5803, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(321.9421, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(483.9438, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(277.3140, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(467.2949, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(504.8327, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(307.7321, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(222.5278, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(193.9592, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(569.1786, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(464.0625, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(196.1591, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(154.0673, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.34017278617710583, 'recall': 0.36885245901639346, 'f1': 0.35393258426966295, 'number': 854}, 'P': {'precision': 0.4181985294117647, 'recall': 0.5462184873949579, 'f1': 0.47371160853722016, 'number': 833}, 'overall_precision': 0.38232373386295926, 'overall_recall': 0.45643153526970953, 'overall_f1': 0.41610375574169145, 'overall_accuracy': 0.6366929170658753}
			------------EPOCH 11---------------
Loss:  tensor(162.8902, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(150.2702, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(199.0444, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(124.6587, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(245.3433, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(364.6060, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(137.6518, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(327.8198, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(217.6819, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(200.3896, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(194.3900, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(243.1409, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(259.6855, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(388.5659, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(217.9149, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(386.9937, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(435.1220, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(377.2211, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(354.5828, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(339.6140, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(221.0117, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.3070, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(228.6226, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(266.3284, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(197.0409, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(339.0145, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(341.1689, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(226.7914, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(133.0363, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(153.0540, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(384.3989, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(333.0624, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(168.6602, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(142.8372, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.322463768115942, 'recall': 0.20843091334894615, 'f1': 0.2532005689900427, 'number': 854}, 'P': {'precision': 0.38935574229691877, 'recall': 0.6674669867947179, 'f1': 0.4918177797434763, 'number': 833}, 'overall_precision': 0.3707070707070707, 'overall_recall': 0.4350918790752816, 'overall_f1': 0.4003272429779111, 'overall_accuracy': 0.6314817844039913}
			------------EPOCH 12---------------
Loss:  tensor(125.8450, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(128.7486, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(141.8924, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(82.3904, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(166.8611, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(233.7256, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(83.7795, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(184.6118, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(128.4244, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(141.6702, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(99.0780, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(169.2925, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(162.3717, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(245.7597, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(129.1660, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(253.0602, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(306.7074, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(276.3509, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(263.8081, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(263.3223, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(157.5107, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.3363, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(152.4290, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(209.7888, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(136.9817, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(270.9465, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(228.5089, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(164.2381, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(78.8212, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(115.9115, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(255.6639, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(201.5598, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(64.5053, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(61.7508, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.38235294117647056, 'recall': 0.27400468384074944, 'f1': 0.31923601637107774, 'number': 854}, 'P': {'precision': 0.4018759018759019, 'recall': 0.6686674669867947, 'f1': 0.5020279405137449, 'number': 833}, 'overall_precision': 0.3958958958958959, 'overall_recall': 0.46887966804979253, 'overall_f1': 0.42930800542740843, 'overall_accuracy': 0.636459233986867}
			------------EPOCH 13---------------
Loss:  tensor(82.5313, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(73.6157, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(86.5894, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.3349, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(118.4040, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(173.5614, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(65.3983, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(126.4982, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(117.0484, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(100.0188, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(77.7058, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(152.9043, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(134.9182, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(207.7790, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(111.5348, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(259.3318, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(239.1659, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(214.6643, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(225.0401, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(218.7348, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(107.5577, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.4734, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(119.4663, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(142.0176, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(101.0361, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(179.9257, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(143.5974, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(100.6692, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(67.9457, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(87.9409, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(166.1404, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(140.3853, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(77.4357, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(72.3639, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.33158813263525305, 'recall': 0.4449648711943794, 'f1': 0.37999999999999995, 'number': 854}, 'P': {'precision': 0.39134615384615384, 'recall': 0.4885954381752701, 'f1': 0.4345969033635878, 'number': 833}, 'overall_precision': 0.36001829826166515, 'overall_recall': 0.46650859513930054, 'overall_f1': 0.40640330493157756, 'overall_accuracy': 0.6369032318369827}
			------------EPOCH 14---------------
Loss:  tensor(69.3382, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.1478, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(86.1580, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(68.8829, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(90.4178, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(151.3160, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.4022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(112.2153, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(100.0688, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(62.2666, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(66.2349, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(111.0661, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(102.2289, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(136.5036, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(87.1170, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(138.9110, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(156.6864, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(139.0421, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(154.1597, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(133.6820, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(77.2626, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.0059, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(94.7814, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(112.4774, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(86.3257, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(129.8281, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(104.5271, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(76.3422, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.8448, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(67.2220, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(123.8918, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(98.3592, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.0749, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.9890, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.38336713995943206, 'recall': 0.22131147540983606, 'f1': 0.2806236080178173, 'number': 854}, 'P': {'precision': 0.3866396761133603, 'recall': 0.687875150060024, 'f1': 0.49503239740820726, 'number': 833}, 'overall_precision': 0.3858227848101266, 'overall_recall': 0.45168938944872555, 'overall_f1': 0.41616602949208087, 'overall_accuracy': 0.6183487953637277}
			------------EPOCH 15---------------
Loss:  tensor(57.1416, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.3984, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(70.9681, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.3414, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(72.2013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(106.4339, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.2624, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(100.6678, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(70.8866, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.5414, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.8875, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(86.7358, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.3005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(87.3679, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.8294, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(100.5063, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(112.2506, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(94.7542, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(121.3932, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(108.8659, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(61.9820, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.3903, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(78.2263, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(96.2234, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(69.8991, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(91.1347, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(76.2051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.4691, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.1841, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.0934, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(85.0182, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(68.3587, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.0923, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.2656, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3675977653631285, 'recall': 0.38524590163934425, 'f1': 0.3762149799885649, 'number': 854}, 'P': {'precision': 0.4005123825789923, 'recall': 0.5630252100840336, 'f1': 0.468063872255489, 'number': 833}, 'overall_precision': 0.3862536302032914, 'overall_recall': 0.4730290456431535, 'overall_f1': 0.42525979216626697, 'overall_accuracy': 0.645269086065478}
			------------EPOCH 16---------------
Loss:  tensor(38.2633, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.2117, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.5500, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.7371, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.5933, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(75.5245, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.7628, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.8301, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(64.7148, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.8367, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.1823, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(66.5326, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.1320, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(74.5863, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.7006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(75.8012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(94.1460, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(86.1812, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(125.0593, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(99.3737, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.1435, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.5915, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(65.4437, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(100.4851, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(61.8179, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(76.0668, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(67.3967, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.2729, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.1815, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.8847, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(79.0049, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(62.4001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.8059, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.1338, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3963838664812239, 'recall': 0.3337236533957845, 'f1': 0.36236490781945324, 'number': 854}, 'P': {'precision': 0.3945371775417299, 'recall': 0.6242496998799519, 'f1': 0.4834960483496048, 'number': 833}, 'overall_precision': 0.3951890034364261, 'overall_recall': 0.47717842323651455, 'overall_f1': 0.4323308270676692, 'overall_accuracy': 0.6418105764961559}
			------------EPOCH 17---------------
Loss:  tensor(33.8985, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.8671, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.4948, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.2215, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.0330, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.4719, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.4769, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.3160, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.0476, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.1694, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.5906, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.2215, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.2352, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.7066, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.0986, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(64.9577, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(77.1101, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(71.4292, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(93.0060, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(98.3414, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.1291, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.0678, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.7185, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(79.2947, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(68.6624, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(73.9721, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.7774, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.4153, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.9728, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.6984, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(68.8591, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(64.9371, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.0532, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.7899, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.413075780089153, 'recall': 0.3255269320843091, 'f1': 0.364112639161755, 'number': 854}, 'P': {'precision': 0.3978021978021978, 'recall': 0.6518607442977191, 'f1': 0.4940855323020929, 'number': 833}, 'overall_precision': 0.4028459273797841, 'overall_recall': 0.4866627148784825, 'overall_f1': 0.4408053691275168, 'overall_accuracy': 0.6434697263571145}
			------------EPOCH 18---------------
Loss:  tensor(39.6607, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.5921, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.4411, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.1094, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.8960, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.8935, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.1834, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.5429, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.6366, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.7025, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.6554, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.0476, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.2017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.8008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.2235, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.7556, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.9167, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.4095, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(74.3606, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(70.9066, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.4113, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.4035, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.8069, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(70.4230, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.8448, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(59.6621, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.5557, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.3990, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(59.8092, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.9265, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.9391, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.2168, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.5598, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.3006, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3778580024067389, 'recall': 0.36768149882903983, 'f1': 0.37270029673590505, 'number': 854}, 'P': {'precision': 0.3946078431372549, 'recall': 0.5798319327731093, 'f1': 0.4696159455517744, 'number': 833}, 'overall_precision': 0.3878345498783455, 'overall_recall': 0.4724362774155305, 'overall_f1': 0.42597541421699625, 'overall_accuracy': 0.6470684457738415}
			------------EPOCH 19---------------
Loss:  tensor(34.6431, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.6846, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.8150, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.5959, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.9299, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.7003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.7612, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.7668, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.1452, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.7152, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.9300, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.7364, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.8339, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.7157, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.5080, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.5620, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.4677, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.2997, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(64.6867, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.5229, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.7860, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.3407, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.0089, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(59.9242, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.0171, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.8921, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.4181, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.4125, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.4010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.9835, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.1243, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.4287, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.4778, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.2240, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.39920948616600793, 'recall': 0.3548009367681499, 'f1': 0.3756974581525109, 'number': 854}, 'P': {'precision': 0.4028103044496487, 'recall': 0.6194477791116446, 'f1': 0.4881740775780511, 'number': 833}, 'overall_precision': 0.40147058823529413, 'overall_recall': 0.4854771784232365, 'overall_f1': 0.4394955728467937, 'overall_accuracy': 0.6476993900871638}
			------------EPOCH 20---------------
Loss:  tensor(29.2015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.8040, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.8574, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.2460, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.0839, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.6686, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.0098, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.2947, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.6961, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.6207, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.5388, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.0369, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.9454, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.0659, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.5456, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.8396, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.7498, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.7181, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.2515, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.2206, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.9358, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.7219, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.0106, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.8340, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.4757, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.6361, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.1831, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.4206, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.6014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.1150, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.5499, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.1698, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.7885, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.8994, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.40827586206896554, 'recall': 0.34660421545667447, 'f1': 0.37492083597213427, 'number': 854}, 'P': {'precision': 0.397125567322239, 'recall': 0.6302521008403361, 'f1': 0.48723897911832953, 'number': 833}, 'overall_precision': 0.40107474352711286, 'overall_recall': 0.4866627148784825, 'overall_f1': 0.4397429030530262, 'overall_accuracy': 0.6419040497277592}
			------------EPOCH 21---------------
Loss:  tensor(25.0022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.7177, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.6727, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.8857, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.4203, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.5204, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.3996, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.8959, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.9910, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.1647, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.2509, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.9493, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.5480, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.9866, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.7681, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.8333, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.5129, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.2636, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.4144, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.5226, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.2503, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.0655, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.2330, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.2932, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.1444, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.9141, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.5099, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.7824, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.1449, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.5820, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.3001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.3205, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.6184, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.2692, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3941411451398136, 'recall': 0.34660421545667447, 'f1': 0.3688473520249222, 'number': 854}, 'P': {'precision': 0.3967766692248657, 'recall': 0.6206482593037215, 'f1': 0.4840823970037453, 'number': 833}, 'overall_precision': 0.3958130477117819, 'overall_recall': 0.4819205690574985, 'overall_f1': 0.4346431435445068, 'overall_accuracy': 0.6404552146379081}
			------------EPOCH 22---------------
Loss:  tensor(23.5758, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.2155, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.6544, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.5347, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.4780, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.2429, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.4719, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.2789, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.3465, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.2077, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.1374, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.4141, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.2417, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.9940, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.2154, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.5218, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.0544, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.4206, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.1219, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.6294, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.9406, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.1328, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.7582, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.0957, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.5774, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.3567, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.0006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.0673, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.0407, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.4014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.0201, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.1350, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.7392, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.5798, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.39011703511053314, 'recall': 0.351288056206089, 'f1': 0.36968576709796674, 'number': 854}, 'P': {'precision': 0.39197530864197533, 'recall': 0.60984393757503, 'f1': 0.4772193518083608, 'number': 833}, 'overall_precision': 0.39128329297820824, 'overall_recall': 0.47895672791938354, 'overall_f1': 0.4307036247334755, 'overall_accuracy': 0.6373238613791975}
			------------EPOCH 23---------------
Loss:  tensor(24.1863, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.7479, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.0530, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.6027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.0911, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.5112, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.5972, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.2144, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.7987, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.3365, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.6705, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.7260, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.1480, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.2966, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.8230, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.9376, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.3452, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.2500, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.1136, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.1454, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.3803, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.4863, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.7453, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.5640, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.3592, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.3378, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.2558, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.9448, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.7645, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.3897, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.0693, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.5631, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.7282, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.8124, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.39972714870395637, 'recall': 0.3430913348946136, 'f1': 0.3692501575299307, 'number': 854}, 'P': {'precision': 0.3862433862433862, 'recall': 0.6134453781512605, 'f1': 0.47402597402597396, 'number': 833}, 'overall_precision': 0.39105058365758755, 'overall_recall': 0.47658565500889155, 'overall_f1': 0.42960192359070265, 'overall_accuracy': 0.6393802724744702}
			------------EPOCH 24---------------
Loss:  tensor(21.5649, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.3624, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.1497, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.1139, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.1349, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.4823, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.6651, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.6038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.9310, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.4598, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.2203, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.9154, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.6632, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.1719, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.2056, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.9149, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.5783, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.5146, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.7673, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.5655, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.6056, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.4056, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.3005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.7332, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.1471, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.4530, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.9709, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.0529, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.4908, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.3667, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.8059, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.7687, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.6064, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.0731, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4041666666666667, 'recall': 0.3407494145199063, 'f1': 0.3697585768742059, 'number': 854}, 'P': {'precision': 0.3914044512663085, 'recall': 0.6122448979591837, 'f1': 0.4775280898876404, 'number': 833}, 'overall_precision': 0.39594661393969355, 'overall_recall': 0.4748073503260225, 'overall_f1': 0.43180592991913747, 'overall_accuracy': 0.6401046900193957}
			------------EPOCH 25---------------
Loss:  tensor(19.9048, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.2056, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.0442, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.1284, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.4119, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.6283, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.0563, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.8258, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.8010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.6402, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.6391, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.8180, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.4845, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.7704, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.2291, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.0233, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.6382, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.9378, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.5095, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.2616, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.5317, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.6436, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.3999, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.4672, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.1092, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.3032, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.3889, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.9734, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.0523, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.2967, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.6142, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.0048, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.6654, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.1541, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.41007194244604317, 'recall': 0.3337236533957845, 'f1': 0.36797934151065204, 'number': 854}, 'P': {'precision': 0.3895612708018154, 'recall': 0.6182472989195679, 'f1': 0.4779582366589327, 'number': 833}, 'overall_precision': 0.3966286564204264, 'overall_recall': 0.4742145820983995, 'overall_f1': 0.4319654427645789, 'overall_accuracy': 0.6398242703245858}
			------------EPOCH 26---------------
Loss:  tensor(19.2712, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.6867, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.0342, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.2473, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.1537, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.8938, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.7927, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.9087, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.2605, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.7219, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.4514, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.2059, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.8488, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.8523, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.5427, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.9348, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.2349, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.6782, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.7505, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.6523, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.5604, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.1955, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.7312, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.3080, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.4473, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.3991, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.2281, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.7080, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.2427, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.8423, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.4702, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.9214, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.4251, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.8718, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.41763005780346824, 'recall': 0.33840749414519905, 'f1': 0.3738680465717982, 'number': 854}, 'P': {'precision': 0.3950151057401813, 'recall': 0.6278511404561825, 'f1': 0.4849327770050997, 'number': 833}, 'overall_precision': 0.4027777777777778, 'overall_recall': 0.48132780082987553, 'overall_f1': 0.43856332703213613, 'overall_accuracy': 0.640198163250999}
			------------EPOCH 27---------------
Loss:  tensor(17.8963, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.3142, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.1229, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.6284, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.9748, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.0315, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.6162, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.2819, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.1228, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.0498, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.0759, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.0422, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.3257, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.3156, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.6286, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.2527, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.4865, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.7571, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.8682, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.0532, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.5226, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.6289, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.3262, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.8678, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.3294, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.9955, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.5587, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.3353, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.4116, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.6233, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.6993, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.7153, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.1570, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.8605, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4122681883024251, 'recall': 0.33840749414519905, 'f1': 0.3717041800643087, 'number': 854}, 'P': {'precision': 0.3854875283446712, 'recall': 0.6122448979591837, 'f1': 0.47309833024118736, 'number': 833}, 'overall_precision': 0.39476284584980237, 'overall_recall': 0.4736218138707765, 'overall_f1': 0.43061169496092694, 'overall_accuracy': 0.638866169700652}
			------------EPOCH 28---------------
Loss:  tensor(16.9265, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.6629, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.3674, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.1166, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.8115, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.3276, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.8449, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.0840, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.9934, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.3990, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.6216, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.7715, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.6216, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.4733, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.0243, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.8347, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.7228, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.6216, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.2649, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.6673, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.4083, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.2862, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.6840, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.8575, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.5395, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.9142, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.8234, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.6778, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.7628, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.7167, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.2721, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.3888, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.8243, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.2056, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4041666666666667, 'recall': 0.3407494145199063, 'f1': 0.3697585768742059, 'number': 854}, 'P': {'precision': 0.39223153084539225, 'recall': 0.6182472989195679, 'f1': 0.4799627213420318, 'number': 833}, 'overall_precision': 0.396458435809149, 'overall_recall': 0.47777119146413755, 'overall_f1': 0.43333333333333335, 'overall_accuracy': 0.6383754352347346}
			------------EPOCH 29---------------
Loss:  tensor(15.0690, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.6792, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.8753, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.6906, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.2715, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.9918, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.3117, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.3906, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.4804, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.8080, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.6058, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.1073, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.3937, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.3870, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.6626, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.0447, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.1679, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.3692, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.4419, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.5071, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.1259, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.3844, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.6843, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.6476, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.2405, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.4501, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.4470, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.3759, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.0611, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.3317, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.6058, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.5765, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.5233, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.0277, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3936899862825789, 'recall': 0.3360655737704918, 'f1': 0.3626026531901453, 'number': 854}, 'P': {'precision': 0.3949962092494314, 'recall': 0.6254501800720288, 'f1': 0.4842007434944238, 'number': 833}, 'overall_precision': 0.39453125, 'overall_recall': 0.47895672791938354, 'overall_f1': 0.43266398929049527, 'overall_accuracy': 0.6344495595073961}
			------------EPOCH 30---------------
Loss:  tensor(14.8261, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.9802, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.5040, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.3693, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.0880, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.3151, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.8101, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.0246, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.1358, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5339, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.8916, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.4938, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.7128, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.4579, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.1627, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.0955, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.5384, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.1421, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.7791, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.4730, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.7767, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.0855, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.6070, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.4927, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.2952, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.5437, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.8375, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.5538, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.5343, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.6173, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.5852, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.6610, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.6573, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.2269, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3851744186046512, 'recall': 0.31030444964871196, 'f1': 0.3437094682230869, 'number': 854}, 'P': {'precision': 0.3945225758697261, 'recall': 0.6398559423769508, 'f1': 0.4880952380952381, 'number': 833}, 'overall_precision': 0.39136831780284453, 'overall_recall': 0.4730290456431535, 'overall_f1': 0.42834138486312395, 'overall_accuracy': 0.6381651204636273}


		-------------RUN 5-----------
			------------EPOCH 1---------------
Loss:  tensor(2833.9053, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1437.9912, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1728.9592, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2487.3914, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1307.5215, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2695.4187, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(587.7489, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1275.3894, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1150.5731, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(993.2019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2043.1897, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1981.2607, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2300.2122, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2787.7830, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2289.5786, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2497.8687, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1513.8636, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2163.0273, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1979.4008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1747.2976, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2906.6904, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2457.0879, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1987.4211, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1640.1704, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1723.8774, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1543.8411, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1660.4594, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2167.7241, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1934.7034, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1310.6965, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1536.0049, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1441.9302, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1304.3760, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1500.9758, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(710.2015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2383.3940, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.17857142857142858, 'recall': 0.07115135834411385, 'f1': 0.10175763182238667, 'number': 773}, 'P': {'precision': 0.27452085682074406, 'recall': 0.5025799793601651, 'f1': 0.35508567262121765, 'number': 969}, 'overall_precision': 0.26032660902977905, 'overall_recall': 0.3111366245694604, 'overall_f1': 0.2834728033472803, 'overall_accuracy': 0.500920326058375}
			------------EPOCH 2---------------
Loss:  tensor(2124.2185, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1034.2903, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1220.6731, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1730.5981, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(919.7512, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1884.8177, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(443.0470, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(992.6362, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(933.9083, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(794.4738, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1631.3259, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1549.9675, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1905.9740, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2371.3552, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1965.2222, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2199.5498, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1212.8501, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1887.0957, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1568.1392, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1396.6754, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2461.4839, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2059.3054, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1546.5591, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1324.1149, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1481.6261, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1253.6331, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1359.7480, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1754.5076, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1607.9026, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1060.3488, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1214.1785, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1239.0200, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1104.8958, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1275.4854, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(554.8641, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2053.7617, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.1523809523809524, 'recall': 0.1034928848641656, 'f1': 0.12326656394453005, 'number': 773}, 'P': {'precision': 0.30781938325991187, 'recall': 0.5768833849329206, 'f1': 0.4014362657091562, 'number': 969}, 'overall_precision': 0.2729602733874413, 'overall_recall': 0.3668197474167623, 'overall_f1': 0.3130051432770022, 'overall_accuracy': 0.540954509597686}
			------------EPOCH 3---------------
Loss:  tensor(1944.3748, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(843.3329, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1002.3918, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1387.5276, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(693.8924, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1446.4535, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(318.2041, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(818.2507, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(821.7175, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(655.8608, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1317.9302, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1289.4634, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1556.0613, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2066.7935, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1721.5574, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1989.2468, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(949.9813, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1543.6221, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1169.4235, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1119.5428, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2051.3701, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1659.6868, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1183.7858, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1067.0199, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1262.1179, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1010.6946, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1081.7808, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1344.7487, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1364.8175, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(848.2031, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(980.9939, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1079.0466, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(911.4612, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1095.2324, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(419.9754, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1747.9910, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.22473118279569892, 'recall': 0.2703751617076326, 'f1': 0.24544920728126834, 'number': 773}, 'P': {'precision': 0.2916939175931982, 'recall': 0.4602683178534572, 'f1': 0.35708566853482787, 'number': 969}, 'overall_precision': 0.26636844245628305, 'overall_recall': 0.3760045924225029, 'overall_f1': 0.3118305165436801, 'overall_accuracy': 0.5640503111578579}
			------------EPOCH 4---------------
Loss:  tensor(1753.3694, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(655.9327, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(734.0160, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1120.2068, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(556.0892, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1148.2500, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(220.0863, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(725.5649, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(726.0674, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(515.1499, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1095.5481, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1135.5880, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1301.2311, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1795.9116, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1451.8184, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1682.2751, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(779.3475, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1267.1626, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(887.0638, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(895.1489, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1738.8245, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1344.8564, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(940.5251, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(910.7879, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1040.4695, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(872.5658, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(894.9734, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1011.1587, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1115.5229, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(713.7837, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(793.1257, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(892.8566, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(685.4675, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(879.9425, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(299.5206, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1462.1071, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.24960254372019078, 'recall': 0.40620957309184996, 'f1': 0.30920728705071393, 'number': 773}, 'P': {'precision': 0.35814722911497104, 'recall': 0.44685242518059853, 'f1': 0.39761248852157943, 'number': 969}, 'overall_precision': 0.30279691933522496, 'overall_recall': 0.4288174512055109, 'overall_f1': 0.35495367070563083, 'overall_accuracy': 0.598935051275309}
			------------EPOCH 5---------------
Loss:  tensor(1339.1989, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(509.7758, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(540.3615, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(842.0739, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(401.0232, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(841.5394, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(181.8569, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(638.7937, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(606.1517, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(401.8642, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(801.1489, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(864.5157, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(978.8665, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1395.2344, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1106.0293, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1266.3556, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(573.5369, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(801.2499, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(611.4202, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(749.7256, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1310.9749, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(920.4390, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(730.8444, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(754.0087, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(752.7114, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(607.8801, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(552.0259, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(671.2156, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(802.6243, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(561.5483, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(612.8099, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(623.3914, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(472.6961, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(675.2744, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(229.2703, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1021.1362, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3205268935236004, 'recall': 0.3777490297542044, 'f1': 0.346793349168646, 'number': 773}, 'P': {'precision': 0.38997214484679665, 'recall': 0.5779153766769866, 'f1': 0.46569646569646567, 'number': 969}, 'overall_precision': 0.3630166169578185, 'overall_recall': 0.4890929965556831, 'overall_f1': 0.4167278063096112, 'overall_accuracy': 0.6133315803313174}
			------------EPOCH 6---------------
Loss:  tensor(841.3309, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(377.5016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(372.8326, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(565.6057, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(205.1821, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(443.5748, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(128.6866, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(466.2577, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(450.6492, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(273.3485, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(490.5061, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(492.9128, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(575.2736, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(949.1355, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(712.6190, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(795.7476, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(517.3420, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(544.8298, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(434.6096, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(601.3212, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1023.2869, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(561.4630, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(537.8298, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(509.5223, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(536.0984, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(404.6931, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(322.9274, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(380.8745, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(479.8646, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(435.9911, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(433.0134, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(472.8000, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(341.8941, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(490.2857, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(167.3073, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(549.7251, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.2977435332966428, 'recall': 0.6998706338939198, 'f1': 0.41776061776061774, 'number': 773}, 'P': {'precision': 0.4629294755877034, 'recall': 0.26418988648090813, 'f1': 0.33639947437582124, 'number': 969}, 'overall_precision': 0.3362869198312236, 'overall_recall': 0.45752009184845005, 'overall_f1': 0.3876459143968872, 'overall_accuracy': 0.5679726531685512}
			------------EPOCH 7---------------
Loss:  tensor(719.3044, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(328.3165, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(354.0721, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(532.8436, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(157.4324, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(338.8321, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(145.8186, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(527.4061, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(342.2426, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(193.0857, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(368.2103, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(382.1462, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(378.3716, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(568.6094, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(461.2063, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(653.2179, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(707.0079, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(632.8793, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(461.4429, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(573.5117, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(572.1409, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(377.4226, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(468.4878, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(355.1833, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(408.9155, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(359.4817, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(271.7043, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(380.9689, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(487.2258, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(402.9450, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(494.1094, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(343.5525, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(228.0657, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(322.8182, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(140.2139, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(403.2207, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3427592116538132, 'recall': 0.517464424320828, 'f1': 0.41237113402061853, 'number': 773}, 'P': {'precision': 0.4797891036906854, 'recall': 0.5634674922600619, 'f1': 0.5182724252491693, 'number': 969}, 'overall_precision': 0.410412147505423, 'overall_recall': 0.5430539609644087, 'overall_f1': 0.46750679515690635, 'overall_accuracy': 0.6479533701463757}
			------------EPOCH 8---------------
Loss:  tensor(305.4170, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(168.4233, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(137.6299, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(256.3945, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(97.3588, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(206.1385, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(77.0826, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(381.1268, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(215.9553, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(121.3232, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(273.3506, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(231.3348, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(233.6022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(370.0551, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(272.5312, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(443.5504, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(199.4791, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(244.4552, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(283.1863, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(230.0251, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(436.7783, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(355.8217, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(316.4390, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(273.0430, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(346.6309, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(182.9350, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(133.0434, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(144.7750, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(319.6752, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(265.1239, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(278.9728, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(323.7413, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(142.6492, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(259.6578, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(122.3296, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(282.1864, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3136574074074074, 'recall': 0.7011642949547219, 'f1': 0.43342662934826065, 'number': 773}, 'P': {'precision': 0.3952569169960474, 'recall': 0.20639834881320948, 'f1': 0.2711864406779661, 'number': 969}, 'overall_precision': 0.3321396598030439, 'overall_recall': 0.425947187141217, 'overall_f1': 0.3732394366197183, 'overall_accuracy': 0.5905644666491366}
			------------EPOCH 9---------------
Loss:  tensor(221.7331, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(123.0190, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(122.4907, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(274.9323, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(105.7673, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(317.4773, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.1059, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(361.6686, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(520.6906, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(341.2404, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(490.5940, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(753.7736, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(224.3640, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(454.6378, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(286.5890, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(373.2668, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(145.7631, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(152.9925, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(197.0344, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(172.9748, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(437.9379, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(325.1949, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(429.8405, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(345.7835, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(540.6079, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(547.9407, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(474.9897, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(488.0588, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(415.1109, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(367.3083, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(391.2573, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(394.8390, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(143.6204, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(315.9041, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(115.1628, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(376.1987, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.30247641509433965, 'recall': 0.6636481241914618, 'f1': 0.41555285540704745, 'number': 773}, 'P': {'precision': 0.3668639053254438, 'recall': 0.25593395252837975, 'f1': 0.30151975683890575, 'number': 969}, 'overall_precision': 0.3208263069139966, 'overall_recall': 0.4368541905855339, 'overall_f1': 0.3699562469615945, 'overall_accuracy': 0.6136383556841091}
			------------EPOCH 10---------------
Loss:  tensor(237.2679, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(162.9519, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(165.7811, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(270.9514, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(126.8996, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(306.2240, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(90.3700, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(394.5089, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(392.7509, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(247.2603, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(655.0620, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(402.5059, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1025.2751, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(996.8304, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(831.3618, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(626.1606, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(792.8951, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(969.9761, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(684.7584, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(747.8159, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(661.5911, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(659.8154, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(462.3772, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(376.1416, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(190.1341, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(164.4991, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(129.9948, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(183.9689, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(267.9540, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(289.5999, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(272.1368, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(254.7449, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(420.1158, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(708.1074, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(281.8296, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(832.2943, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4782608695652174, 'recall': 0.12807244501940493, 'f1': 0.20204081632653062, 'number': 773}, 'P': {'precision': 0.3607235142118863, 'recall': 0.7203302373581011, 'f1': 0.4807162534435262, 'number': 969}, 'overall_precision': 0.37208216619981327, 'overall_recall': 0.45752009184845005, 'overall_f1': 0.41040164778578786, 'overall_accuracy': 0.5990884389517048}
			------------EPOCH 11---------------
Loss:  tensor(1004.7147, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(769.2665, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(341.3591, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(927.7838, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(374.8123, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(789.0961, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(119.3437, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(724.2911, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(333.6615, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(363.6941, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(627.7672, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(479.1615, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(364.5731, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(395.8519, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(265.6359, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(530.7207, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(204.9199, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(181.8980, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(236.7985, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(228.2245, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(659.2518, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(455.8037, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(295.1450, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(307.8887, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(419.4974, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(463.7273, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(314.7849, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(438.1328, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(474.4614, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(299.1046, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(358.3336, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(387.4707, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(281.2371, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(295.9740, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(183.0341, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(541.8555, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3224789915966387, 'recall': 0.39715394566623546, 'f1': 0.3559420289855073, 'number': 773}, 'P': {'precision': 0.39362912400455063, 'recall': 0.3570691434468524, 'f1': 0.37445887445887444, 'number': 969}, 'overall_precision': 0.3566357181867832, 'overall_recall': 0.3748564867967853, 'overall_f1': 0.36551917156451164, 'overall_accuracy': 0.6321982645280042}
			------------EPOCH 12---------------
Loss:  tensor(337.3716, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(210.3243, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(165.9277, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(359.0803, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(100.3548, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(222.0865, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.3238, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(285.1740, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(177.8294, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(79.9778, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(228.4495, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(194.9669, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(191.4449, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(246.8059, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(194.9106, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(285.4329, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(164.5906, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(147.4148, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(173.4549, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(189.1897, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(270.1673, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(260.4427, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(204.1944, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(176.1588, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(156.0317, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(140.4937, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(112.7937, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(154.9220, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(210.5535, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(230.0459, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(217.8585, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(193.8956, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(236.4327, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(293.9981, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(118.0093, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(392.0968, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3835616438356164, 'recall': 0.47089262613195343, 'f1': 0.4227642276422764, 'number': 773}, 'P': {'precision': 0.4585714285714286, 'recall': 0.6625386996904025, 'f1': 0.5420008442380752, 'number': 969}, 'overall_precision': 0.4282673478075777, 'overall_recall': 0.5774971297359357, 'overall_f1': 0.491811293082376, 'overall_accuracy': 0.6512402489262863}
			------------EPOCH 13---------------
Loss:  tensor(381.0721, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(185.4410, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(184.2472, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(377.6723, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(64.1729, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(145.1068, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.5502, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(176.0432, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(133.4991, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.2323, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(132.7768, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(112.5965, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(98.9353, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(152.5573, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(103.9520, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(201.3974, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(99.4194, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(113.2492, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(186.9710, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(146.9223, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(216.6019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(217.4428, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(153.1172, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(189.7558, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(122.0748, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(180.5087, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(88.4268, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(94.1366, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(132.7911, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(136.9989, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(111.2788, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(100.2038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(110.5720, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(123.9321, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(64.5734, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(117.4958, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4190064794816415, 'recall': 0.5019404915912031, 'f1': 0.45673925838728663, 'number': 773}, 'P': {'precision': 0.4930191972076789, 'recall': 0.5830753353973168, 'f1': 0.5342789598108747, 'number': 969}, 'overall_precision': 0.45994208494208494, 'overall_recall': 0.5470723306544202, 'overall_f1': 0.4997378080755113, 'overall_accuracy': 0.6967306512402489}
			------------EPOCH 14---------------
Loss:  tensor(78.8231, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(66.1230, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.9095, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(134.0056, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.8480, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(83.2731, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.6325, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(98.0664, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(96.8148, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.3147, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(104.0621, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(73.4557, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(62.6119, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(94.5222, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(74.8645, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(140.1505, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.5016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(70.8940, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(66.2069, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(76.4315, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(136.5126, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(160.0227, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(91.6458, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(101.7931, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(75.0859, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(93.6122, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.8536, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.4876, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(96.9594, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(92.3675, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(75.1854, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(59.2278, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.3352, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(74.3832, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.9802, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(61.6005, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4292185730464326, 'recall': 0.49029754204398446, 'f1': 0.45772946859903385, 'number': 773}, 'P': {'precision': 0.47213622291021673, 'recall': 0.6295149638802889, 'f1': 0.5395842547545334, 'number': 969}, 'overall_precision': 0.45471264367816094, 'overall_recall': 0.5677382319173364, 'overall_f1': 0.5049782997191729, 'overall_accuracy': 0.6788719432027347}
			------------EPOCH 15---------------
Loss:  tensor(57.1772, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.9465, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.6883, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(79.7920, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.5496, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(61.7146, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.1056, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.2150, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(76.2297, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.7461, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.9560, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.8362, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.6488, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(78.5189, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.2344, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(90.0781, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.5444, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.2919, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.4919, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.2291, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(110.9179, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(136.2433, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(68.0425, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(77.8927, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(59.9682, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(74.1371, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.8953, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.9654, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(76.7027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.0241, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.9099, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.0935, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.1491, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.3926, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.5052, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.4010, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4425287356321839, 'recall': 0.49805950840879687, 'f1': 0.46865489957395007, 'number': 773}, 'P': {'precision': 0.4628751974723539, 'recall': 0.6047471620227038, 'f1': 0.5243847874720358, 'number': 969}, 'overall_precision': 0.4545880149812734, 'overall_recall': 0.5574052812858783, 'overall_f1': 0.5007735946364106, 'overall_accuracy': 0.6894995179244456}
			------------EPOCH 16---------------
Loss:  tensor(40.6639, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.3877, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.2228, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(59.0521, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.0048, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.8042, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.7976, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.2585, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.1686, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.5478, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.7984, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.7771, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.1470, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.7746, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.1093, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.7835, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.6311, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.0387, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.9354, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.6022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(86.6859, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(105.8505, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.7707, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(69.0538, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.6499, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.0206, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.4878, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.3910, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.7121, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.0124, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.4953, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.7592, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.1094, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.6396, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.6196, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.6071, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.42551020408163265, 'recall': 0.5394566623544631, 'f1': 0.4757558471192242, 'number': 773}, 'P': {'precision': 0.47404255319148936, 'recall': 0.5748194014447885, 'f1': 0.519589552238806, 'number': 969}, 'overall_precision': 0.4519721577726218, 'overall_recall': 0.5591274397244547, 'overall_f1': 0.4998716961765461, 'overall_accuracy': 0.6883162415636778}
			------------EPOCH 17---------------
Loss:  tensor(34.2238, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.9325, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.6774, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.1595, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.2845, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.3133, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.0215, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.6729, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.1287, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.4256, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.6228, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.2275, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.0648, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.2215, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.6210, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.5177, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.7787, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.9763, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.5176, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.0577, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(75.2921, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(91.8685, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.0042, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(62.8714, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.2052, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.7822, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.9138, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.8359, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.5158, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.7340, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.8014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.8835, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.6304, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.0977, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.6552, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.5672, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.44541484716157204, 'recall': 0.5278137128072445, 'f1': 0.48312611012433393, 'number': 773}, 'P': {'precision': 0.47710241465445463, 'recall': 0.5913312693498453, 'f1': 0.528110599078341, 'number': 969}, 'overall_precision': 0.4633915918752952, 'overall_recall': 0.5631458094144661, 'overall_f1': 0.5084218709510236, 'overall_accuracy': 0.6918879831711806}
			------------EPOCH 18---------------
Loss:  tensor(28.4836, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.9457, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.0494, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.9691, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.0577, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.6096, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.0483, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.7186, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.0699, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.2051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.6730, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.6100, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.3179, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.2612, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.2327, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.4245, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.6009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.9786, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.9377, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.5986, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.6715, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(80.6681, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.2718, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.3488, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.9375, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.7015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.8494, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.3570, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.4159, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.9118, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.9576, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.9587, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.1801, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.2718, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.8256, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.8419, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4483870967741935, 'recall': 0.5394566623544631, 'f1': 0.48972401644157365, 'number': 773}, 'P': {'precision': 0.47381546134663344, 'recall': 0.5882352941176471, 'f1': 0.5248618784530386, 'number': 969}, 'overall_precision': 0.46272855133614627, 'overall_recall': 0.5665901262916189, 'overall_f1': 0.5094193548387097, 'overall_accuracy': 0.6917345954947848}
			------------EPOCH 19---------------
Loss:  tensor(26.5868, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.9436, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.5055, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.8030, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.0903, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.4643, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.4424, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.7346, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.9061, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.3684, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.2866, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.2100, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.6885, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.6327, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.1355, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.4002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.6506, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.1880, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.2445, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.8907, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.3006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(74.0382, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.9684, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.6175, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.6209, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.1474, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.2779, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.8499, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.4735, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.5388, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.6265, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.3206, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.1430, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.2655, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.3068, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.9211, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4384133611691023, 'recall': 0.5433376455368694, 'f1': 0.4852686308492201, 'number': 773}, 'P': {'precision': 0.47711864406779664, 'recall': 0.5810113519091847, 'f1': 0.5239646347138204, 'number': 969}, 'overall_precision': 0.4597754911131899, 'overall_recall': 0.5642939150401837, 'overall_f1': 0.506701030927835, 'overall_accuracy': 0.6886668419668682}
			------------EPOCH 20---------------
Loss:  tensor(23.7729, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.8540, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.9922, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.2594, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.3522, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.2514, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.5277, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.8056, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.3050, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.1065, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.9606, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.7518, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.4412, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.5243, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.1721, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.0732, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.0427, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.9709, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.4273, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.6345, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.6199, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(67.5239, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.9518, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.1573, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.0413, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.9908, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.8932, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.7043, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.0853, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.9589, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.9213, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.8540, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.1410, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.3720, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.3630, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.1369, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4430512016718913, 'recall': 0.5485122897800776, 'f1': 0.4901734104046243, 'number': 773}, 'P': {'precision': 0.4794520547945205, 'recall': 0.5779153766769866, 'f1': 0.5240992044922789, 'number': 969}, 'overall_precision': 0.46305882352941174, 'overall_recall': 0.5648679678530425, 'overall_f1': 0.508921644685803, 'overall_accuracy': 0.6897186431764396}
			------------EPOCH 21---------------
Loss:  tensor(22.1686, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.4213, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.0390, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.7408, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.1463, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.0177, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.3622, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.0144, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.1751, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.1332, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.2943, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.0166, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.2677, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.7781, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.3134, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.7822, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.7997, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.5030, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.1573, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.7814, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.9751, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(62.3942, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.2985, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.0805, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.7707, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.2171, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.2938, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.0428, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.7146, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.7995, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.8612, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.6079, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.6550, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.7778, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.1324, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.7563, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4459601259181532, 'recall': 0.5498059508408797, 'f1': 0.492468134414832, 'number': 773}, 'P': {'precision': 0.4901117798796217, 'recall': 0.5882352941176471, 'f1': 0.5347091932457787, 'number': 969}, 'overall_precision': 0.47022684310018903, 'overall_recall': 0.571182548794489, 'overall_f1': 0.5158113011923277, 'overall_accuracy': 0.6901788062056271}
			------------EPOCH 22---------------
Loss:  tensor(20.5403, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.2177, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.1654, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.8755, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.2669, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.9140, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.4394, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.5422, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.6530, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.4066, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.8566, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.0503, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.2675, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.2993, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.6495, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.0068, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.6889, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.1113, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.3669, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.5376, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.5435, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.5472, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.7299, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.8100, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.4457, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.1325, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.9301, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.6249, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.2263, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.9909, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.1281, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.4322, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.0196, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.6342, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.9359, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.4097, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4437434279705573, 'recall': 0.5459249676584734, 'f1': 0.4895591647331787, 'number': 773}, 'P': {'precision': 0.4896373056994819, 'recall': 0.5851393188854489, 'f1': 0.5331452750352609, 'number': 969}, 'overall_precision': 0.46894262683736365, 'overall_recall': 0.5677382319173364, 'overall_f1': 0.5136328226434692, 'overall_accuracy': 0.690880007012008}
			------------EPOCH 23---------------
Loss:  tensor(19.3137, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.2861, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.4382, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.1577, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.3350, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.0912, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.5153, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.3522, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.3271, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.7539, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.6304, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.4021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.9790, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.7324, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.0608, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.7956, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.7286, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.2667, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.9295, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.0295, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.0537, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.4567, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.6011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.7488, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.2491, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.6573, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.8538, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.4944, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.4116, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.3073, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.8261, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.3154, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.4938, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.1726, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.2481, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.3910, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.43302180685358255, 'recall': 0.5394566623544631, 'f1': 0.48041474654377875, 'number': 773}, 'P': {'precision': 0.48664944013781225, 'recall': 0.5830753353973168, 'f1': 0.5305164319248826, 'number': 969}, 'overall_precision': 0.4623352165725047, 'overall_recall': 0.563719862227325, 'overall_f1': 0.5080186239006725, 'overall_accuracy': 0.6899596809536331}
			------------EPOCH 24---------------
Loss:  tensor(18.2558, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.3380, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.8313, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.7259, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.4229, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.6210, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.6519, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.1616, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.3771, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.1821, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.5382, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.0479, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.9929, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.4783, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.7752, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.6694, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.9883, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.5367, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.6400, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.7340, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.9083, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.0981, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.4134, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.5765, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.0200, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.5334, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.8644, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.3311, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.8623, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.5044, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.7030, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.2773, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.5651, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.3463, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.6301, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.2868, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.43685300207039335, 'recall': 0.5459249676584734, 'f1': 0.48533640023001723, 'number': 773}, 'P': {'precision': 0.4840379637618637, 'recall': 0.5789473684210527, 'f1': 0.5272556390977444, 'number': 969}, 'overall_precision': 0.46258823529411763, 'overall_recall': 0.5642939150401837, 'overall_f1': 0.508404447892423, 'overall_accuracy': 0.6906827942852134}
			------------EPOCH 25---------------
Loss:  tensor(17.2774, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.5656, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.2061, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.5804, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.7546, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.4081, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.9647, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.2550, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.5327, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.7734, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.5657, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.0744, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.3220, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.9908, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.7686, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.7510, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.2392, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.3504, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.5100, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.8075, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.0988, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.8727, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.3350, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.7031, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.1256, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.5035, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.8569, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.8796, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.3784, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.1510, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.7081, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.4589, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.6996, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.5758, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.1399, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.1625, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4375644994840041, 'recall': 0.5485122897800776, 'f1': 0.48679678530424797, 'number': 773}, 'P': {'precision': 0.48879310344827587, 'recall': 0.5851393188854489, 'f1': 0.5326444340065759, 'number': 969}, 'overall_precision': 0.46547674964772195, 'overall_recall': 0.568886337543054, 'overall_f1': 0.5120123998966676, 'overall_accuracy': 0.6896090805504427}
			------------EPOCH 26---------------
Loss:  tensor(16.4816, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.9238, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.5654, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.4930, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.1663, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.4203, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.4872, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.6289, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.9107, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.3619, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.7012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.5555, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.2709, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.9246, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.9249, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.9623, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.5804, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.6093, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.4972, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.0450, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.2642, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.8044, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.2278, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.3814, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.7089, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.6528, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.9585, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.7954, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.3179, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.2865, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.8714, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.5092, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.8774, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.9217, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.6286, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.1705, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4459601259181532, 'recall': 0.5498059508408797, 'f1': 0.492468134414832, 'number': 773}, 'P': {'precision': 0.4884318766066838, 'recall': 0.5882352941176471, 'f1': 0.5337078651685393, 'number': 969}, 'overall_precision': 0.4693396226415094, 'overall_recall': 0.571182548794489, 'overall_f1': 0.5152770585189022, 'overall_accuracy': 0.6908361819616092}
			------------EPOCH 27---------------
Loss:  tensor(15.7864, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.2425, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.1420, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.6085, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.5522, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.4732, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.0067, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.7721, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.4494, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.8940, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.9545, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.4909, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.4563, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.9849, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.1896, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.2942, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.0389, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.0702, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.5562, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.3168, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.9078, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.1842, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.0883, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.2968, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.6066, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.9537, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.1165, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.9566, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.5123, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.6834, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.1444, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.6627, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.0829, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.3117, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.1694, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.2709, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.44654088050314467, 'recall': 0.5510996119016818, 'f1': 0.493341053850608, 'number': 773}, 'P': {'precision': 0.48717948717948717, 'recall': 0.5882352941176471, 'f1': 0.5329593267882188, 'number': 969}, 'overall_precision': 0.4689265536723164, 'overall_recall': 0.5717566016073479, 'overall_f1': 0.5152612519399897, 'overall_accuracy': 0.6907923569112104}
			------------EPOCH 28---------------
Loss:  tensor(15.2050, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.7728, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.7755, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.8921, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.0576, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.8257, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.6244, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.2243, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.9188, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.5798, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.3670, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.7017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.7060, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.1921, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.5600, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.7481, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.4665, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.3955, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.7291, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.7644, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.1345, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.7973, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.7282, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.3425, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.4379, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.2511, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.2338, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.1529, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.8474, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.2107, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.5175, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.9492, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.3386, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.7624, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.7513, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.5031, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.44607329842931936, 'recall': 0.5510996119016818, 'f1': 0.4930555555555555, 'number': 773}, 'P': {'precision': 0.49101796407185627, 'recall': 0.5923632610939112, 'f1': 0.5369504209541627, 'number': 969}, 'overall_precision': 0.4708097928436911, 'overall_recall': 0.574052812858783, 'overall_f1': 0.5173305742369374, 'overall_accuracy': 0.6910991322640021}
			------------EPOCH 29---------------
Loss:  tensor(14.6275, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.3044, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.5066, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.1577, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.5976, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.2399, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.2968, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.6330, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.5345, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.2139, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.8276, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.0871, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.9718, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.3735, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.9279, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.2320, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.8914, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.9189, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.0326, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.0724, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.4918, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.0039, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.6510, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.5416, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.0136, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.6207, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.4518, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.5629, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.2777, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.8138, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.0049, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.3361, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.5870, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.2772, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.3977, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.8736, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4474789915966387, 'recall': 0.5510996119016818, 'f1': 0.4939130434782609, 'number': 773}, 'P': {'precision': 0.48890784982935154, 'recall': 0.5913312693498453, 'f1': 0.5352638953759925, 'number': 969}, 'overall_precision': 0.4703389830508475, 'overall_recall': 0.5734787600459242, 'overall_f1': 0.5168132436627005, 'overall_accuracy': 0.6912963449907967}
			------------EPOCH 30---------------
Loss:  tensor(14.1486, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.9502, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.2661, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.3994, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.2167, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.7365, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.0708, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.2812, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.1831, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.9035, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.3642, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.5865, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.2797, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.6981, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.3775, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.7973, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.3319, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.3715, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.4270, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.5604, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.1959, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.3638, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.8791, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.8595, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.8383, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.9212, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.7847, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.9471, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.7829, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.4676, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.5538, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.8117, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.8574, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.8478, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.1604, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.3133, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.45052631578947366, 'recall': 0.553686934023286, 'f1': 0.4968078932095183, 'number': 773}, 'P': {'precision': 0.4892887746358183, 'recall': 0.5892672858617131, 'f1': 0.5346441947565543, 'number': 969}, 'overall_precision': 0.4718941898913557, 'overall_recall': 0.5734787600459242, 'overall_f1': 0.5177507126198497, 'overall_accuracy': 0.6916469453939872}
