Tokenizer: allenai/longformer-base-4096 Model: allenai/longformer-base-4096
	Train size: 80 Test size: 20


		-------------RUN 1-----------
			------------EPOCH 1---------------
Loss:  tensor(1508.1382, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3049.2246, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3722.5962, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2583.3738, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1477.6609, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3569.8733, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2199.6448, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2576.1899, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1875.0886, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1620.5107, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2497.8237, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2124.5386, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2644.4595, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2021.7296, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3399.3442, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2153.3306, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1450.6328, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2056.5430, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1037.7040, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1565.2952, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1831.6995, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1398.7426, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1561.8241, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1548.1099, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1755.0638, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(786.5566, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2452.0120, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2030.2104, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(642.3708, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1267.7251, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1104.1101, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(780.0887, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(813.9902, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2242.7156, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2466.4578, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(952.6354, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1172.9211, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2443.1365, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1766.5968, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1669.1571, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1813.6864, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1257.1750, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2582.2563, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1342.9553, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1535.9893, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1132.6426, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1597.3760, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1164.0823, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1589.7540, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(970.0474, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1674.9143, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(900.4512, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2087.6223, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2133.3140, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(819.9714, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1698.1985, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.10585585585585586, 'recall': 0.16666666666666666, 'f1': 0.12947658402203854, 'number': 282}, 'P': {'precision': 0.10270270270270271, 'recall': 0.07279693486590039, 'f1': 0.08520179372197309, 'number': 261}, 'overall_precision': 0.10492845786963434, 'overall_recall': 0.12154696132596685, 'overall_f1': 0.11262798634812286, 'overall_accuracy': 0.48138578978273633}
			------------EPOCH 2---------------
Loss:  tensor(1075.8430, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1887.2095, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2500.4395, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2045.4242, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(955.8031, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2106.8616, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1332.5027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1871.4224, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1355.3701, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1202.6021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1947.5037, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1411.7836, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1719.7112, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1387.2871, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2595.6089, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1597.7539, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1047.3586, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1509.0349, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(732.9014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1077.7205, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1432.0327, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1057.7412, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1261.6683, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1138.7295, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1360.1130, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(595.1403, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2011.5988, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1646.0986, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(527.2740, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(986.7593, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(916.0804, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(623.2593, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(631.3367, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2092.6890, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2178.0093, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(722.9543, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(978.8703, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1910.2557, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1232.5642, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1267.4059, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1445.2034, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(959.8406, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2178.6099, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(999.1237, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1215.5010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(914.9772, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1280.1661, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(954.2889, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1261.1838, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(783.5181, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1425.8621, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(816.2209, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1768.4165, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1662.6711, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(630.3656, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1349.6523, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.13333333333333333, 'recall': 0.14184397163120568, 'f1': 0.1374570446735395, 'number': 282}, 'P': {'precision': 0.218562874251497, 'recall': 0.2796934865900383, 'f1': 0.2453781512605042, 'number': 261}, 'overall_precision': 0.17823343848580442, 'overall_recall': 0.20810313075506445, 'overall_f1': 0.19201359388275277, 'overall_accuracy': 0.5331180270111567}
			------------EPOCH 3---------------
Loss:  tensor(896.0208, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1628.1152, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2067.7153, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1747.8679, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(712.9227, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1532.4084, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1130.5631, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1639.6907, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1110.2249, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(983.3109, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1535.9045, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1024.8755, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1339.1646, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1099.3179, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2019.4469, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1177.4397, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(816.5053, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1150.5375, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(543.6096, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(879.9534, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1175.7705, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(890.8277, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(995.2710, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(933.4653, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1111.6755, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(482.2167, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1716.7156, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1368.7998, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(430.1392, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(797.5272, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(701.6942, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(516.4000, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(518.0597, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1740.8993, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1854.4677, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(558.0426, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(755.2874, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1374.8944, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(679.8956, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1053.7029, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1137.6575, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(644.4122, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1758.9902, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(733.7388, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(852.6343, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(719.9855, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(998.7867, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(789.2241, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(948.8408, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(618.5629, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1158.7966, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(589.9172, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1374.5135, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1330.5603, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(469.8257, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1047.2808, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.28205128205128205, 'recall': 0.1950354609929078, 'f1': 0.23060796645702306, 'number': 282}, 'P': {'precision': 0.26258992805755393, 'recall': 0.2796934865900383, 'f1': 0.2708719851576994, 'number': 261}, 'overall_precision': 0.27061310782241016, 'overall_recall': 0.23572744014732966, 'overall_f1': 0.2519685039370079, 'overall_accuracy': 0.5642982971227246}
			------------EPOCH 4---------------
Loss:  tensor(782.6279, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1416.1982, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1715.2590, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1209.4846, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(496.7921, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1150.2725, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(877.8180, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1281.3958, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(953.0363, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(741.6562, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1208.8518, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(684.6405, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1140.9148, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(825.7881, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1549.9286, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(900.8983, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(558.9125, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(942.5056, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(414.9631, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(631.0938, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(893.5930, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(610.1923, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(767.5739, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(672.4920, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(795.9010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(354.4150, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1303.3632, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1166.9937, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(367.9584, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(640.2672, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(443.0659, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(400.1978, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(355.4596, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1306.8906, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1482.5029, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(496.2022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(568.8640, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1040.3765, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(478.5902, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(764.6808, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1063.7013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(499.8615, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1489.0980, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(512.9976, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(618.1932, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(543.0198, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(707.1108, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(616.8853, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(791.5979, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(526.1705, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(972.0045, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(433.1985, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1283.5598, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1040.9872, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(353.1372, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(842.1641, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.40711462450592883, 'recall': 0.7304964539007093, 'f1': 0.5228426395939085, 'number': 282}, 'P': {'precision': 0.4276729559748428, 'recall': 0.26053639846743293, 'f1': 0.3238095238095238, 'number': 261}, 'overall_precision': 0.4120300751879699, 'overall_recall': 0.5046040515653776, 'overall_f1': 0.45364238410596025, 'overall_accuracy': 0.5701702877275396}
			------------EPOCH 5---------------
Loss:  tensor(638.2151, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(999.1715, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1525.1230, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1176.4332, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(344.6592, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(945.0155, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(619.8274, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(970.3096, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(757.5756, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(563.6583, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(812.8740, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(533.9656, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1198.2516, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(679.4485, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1435.2886, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(957.2950, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(510.9522, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(972.7521, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(382.0878, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(810.4457, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(734.2622, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(729.9964, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(731.1743, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(694.9385, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(591.7600, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(301.2227, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1056.5615, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(911.7462, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(287.7089, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(447.8690, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(282.1812, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(267.6797, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(365.3388, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(856.1259, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1220.8813, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(454.9537, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(765.6925, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1397.0955, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(687.8485, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(703.3168, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1066.5681, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(502.7578, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1347.0642, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(489.3937, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(443.1306, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(437.3381, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(633.0302, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(458.3923, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(709.4070, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(416.1410, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(811.4993, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(369.6019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1138.9675, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(997.6619, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(426.7622, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(890.1664, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4658385093167702, 'recall': 0.26595744680851063, 'f1': 0.33860045146726864, 'number': 282}, 'P': {'precision': 0.3602251407129456, 'recall': 0.735632183908046, 'f1': 0.48362720403022674, 'number': 261}, 'overall_precision': 0.38472622478386165, 'overall_recall': 0.49171270718232046, 'overall_f1': 0.4316895715440582, 'overall_accuracy': 0.5413388138578978}
			------------EPOCH 6---------------
Loss:  tensor(579.7022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1224.7219, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1636.7214, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1348.0347, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(351.4044, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(793.0699, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(657.0859, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1049.8247, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(559.9590, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(524.2745, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(796.5083, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(482.1242, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(866.0176, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(603.0668, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1446.7620, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(671.8526, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(595.3080, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(636.6403, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(300.8206, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(606.3243, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(589.4771, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(496.5479, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(466.1121, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(436.4997, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(489.0105, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(222.8696, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(658.0944, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(719.8248, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(227.0475, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(419.3586, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(218.2085, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(200.6741, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(251.2621, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(946.6595, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(883.6743, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(304.7820, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(399.4933, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(751.2841, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(289.1981, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(513.9468, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(830.0915, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(259.2239, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(915.1151, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(297.6087, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(357.8402, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(428.3641, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(407.5997, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(331.5527, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(647.6672, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(322.1417, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(654.7280, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(309.3340, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(774.9335, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(579.7391, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(233.5421, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(558.6909, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.48672566371681414, 'recall': 0.5851063829787234, 'f1': 0.5314009661835749, 'number': 282}, 'P': {'precision': 0.4981549815498155, 'recall': 0.5172413793103449, 'f1': 0.5075187969924813, 'number': 261}, 'overall_precision': 0.4918032786885246, 'overall_recall': 0.5524861878453039, 'overall_f1': 0.5203816131830009, 'overall_accuracy': 0.6651203758073987}
			------------EPOCH 7---------------
Loss:  tensor(291.6222, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(559.5538, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(814.2961, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(583.6772, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(151.3606, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(436.9777, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(344.3923, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(543.6862, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(410.3574, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(384.1471, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(510.4263, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(359.2029, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(651.5535, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(425.6617, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(734.6654, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(450.2467, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(288.9285, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(461.4833, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(174.7279, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(352.3608, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(276.0684, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(327.3362, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(251.9329, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(309.0318, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(365.5322, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(146.1792, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(464.0738, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(487.1180, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(169.8071, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(424.0884, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(148.0732, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(149.7296, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(156.4732, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(725.4229, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(523.1423, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(210.7582, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(219.0324, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(477.2538, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(165.0859, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(175.5142, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(564.9705, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(139.7849, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(504.4320, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(176.1588, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(182.8688, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(213.2186, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(265.7731, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(263.1938, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(430.9728, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(175.0794, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(372.2451, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(187.4454, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(491.9754, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(367.4023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(149.9774, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(296.7123, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5064102564102564, 'recall': 0.5602836879432624, 'f1': 0.531986531986532, 'number': 282}, 'P': {'precision': 0.5198776758409785, 'recall': 0.6513409961685823, 'f1': 0.5782312925170068, 'number': 261}, 'overall_precision': 0.513302034428795, 'overall_recall': 0.6040515653775322, 'overall_f1': 0.5549915397631133, 'overall_accuracy': 0.6705813270698767}
			------------EPOCH 8---------------
Loss:  tensor(189.4804, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(339.9161, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(540.4335, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(338.8036, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(109.5364, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(212.6995, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(224.5650, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(354.3557, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(260.4419, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(209.3161, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(292.2075, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(194.7400, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(391.7075, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(251.0508, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(402.7743, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(215.2397, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(174.3351, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(259.8192, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(89.5325, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(218.0751, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(131.9403, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(217.8965, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(166.4719, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(157.7914, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(243.0682, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(94.0913, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(208.2030, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(354.0432, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(99.6425, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(197.0684, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.6855, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(106.6063, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(104.7250, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(382.5860, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(312.9054, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(92.8231, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(81.7779, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(250.7684, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(86.3555, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(95.6301, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(491.7843, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(113.0544, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(278.0613, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(123.2692, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(127.1194, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(157.6200, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(189.1365, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(243.9906, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(296.1747, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(110.5550, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(231.7655, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(141.6909, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(259.1780, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(257.4543, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(91.0668, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(152.3470, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.45689655172413796, 'recall': 0.75177304964539, 'f1': 0.5683646112600537, 'number': 282}, 'P': {'precision': 0.5212765957446809, 'recall': 0.37547892720306514, 'f1': 0.4365256124721604, 'number': 261}, 'overall_precision': 0.4754601226993865, 'overall_recall': 0.570902394106814, 'overall_f1': 0.5188284518828452, 'overall_accuracy': 0.625778038755138}
			------------EPOCH 9---------------
Loss:  tensor(186.1850, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(249.6844, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(434.1002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(258.8503, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(64.8189, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(130.1369, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(125.8152, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(268.3532, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(185.5598, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(155.5371, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(254.3073, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(144.3843, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(264.3310, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(184.6407, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(281.0956, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(170.9602, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(150.8701, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(247.1952, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(72.1638, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(146.2128, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(227.9287, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(190.6302, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(133.5202, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(200.1548, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(262.9844, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(85.1210, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(123.7673, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(236.9714, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(80.9708, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(145.0951, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.5653, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(89.9907, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(110.2316, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(305.8655, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(202.3065, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.7885, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(96.9588, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(242.2311, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(116.2641, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(81.0547, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(445.5056, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(94.1345, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(214.8128, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(139.8071, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(84.3947, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(158.5813, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(146.4706, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(120.4320, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(176.2566, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(155.6228, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(160.1854, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(93.1929, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(323.3784, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(218.5999, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(109.3027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(158.8411, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.49693251533742333, 'recall': 0.574468085106383, 'f1': 0.5328947368421053, 'number': 282}, 'P': {'precision': 0.46938775510204084, 'recall': 0.6168582375478927, 'f1': 0.533112582781457, 'number': 261}, 'overall_precision': 0.4828101644245142, 'overall_recall': 0.5948434622467772, 'overall_f1': 0.533003300330033, 'overall_accuracy': 0.6676453317674692}
			------------EPOCH 10---------------
Loss:  tensor(72.2262, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(144.2866, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(218.2087, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(210.6818, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(67.1747, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(101.3347, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.8245, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(145.2633, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(213.3448, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(172.0542, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(237.0183, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(171.9232, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(448.8044, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(228.3988, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(394.7155, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(122.1320, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(129.0191, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(168.2780, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.8820, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(74.7632, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(73.1194, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(71.6389, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(105.0257, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(78.7089, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(167.0251, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(92.8075, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(144.1977, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(120.3482, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(109.8595, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(275.5782, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(75.5271, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(97.8896, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(95.2581, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(449.9012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(269.4183, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(101.4485, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.3279, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(211.1377, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.4349, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.8034, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(326.6335, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(67.5281, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(101.9233, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.3423, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.2096, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(67.9141, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(119.8467, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(122.6546, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(287.7373, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(119.7372, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(310.2702, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(131.9610, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(354.5992, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(238.3901, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(120.8037, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(171.5764, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.47750865051903113, 'recall': 0.48936170212765956, 'f1': 0.48336252189141854, 'number': 282}, 'P': {'precision': 0.506993006993007, 'recall': 0.5555555555555556, 'f1': 0.530164533820841, 'number': 261}, 'overall_precision': 0.49217391304347824, 'overall_recall': 0.5211786372007366, 'overall_f1': 0.5062611806797853, 'overall_accuracy': 0.6677040516735173}
			------------EPOCH 11---------------
Loss:  tensor(54.6419, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(178.1434, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(191.7120, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(128.7103, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.3765, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(62.5611, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(152.0116, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(167.9837, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(82.4760, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(107.6889, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(188.4710, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(77.3603, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(251.8832, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(146.6116, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(108.6135, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(72.2810, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(105.3165, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(193.5833, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(72.9947, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.5700, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(71.4938, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(98.2901, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(173.6718, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(99.7962, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(161.6657, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(101.9395, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(184.5559, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(275.5254, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(78.1869, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(122.8637, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(61.8976, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(65.1008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.4533, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(124.1168, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(138.2652, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.8220, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.9923, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(126.9356, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.2411, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.9604, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(325.5766, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(116.5610, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(87.8249, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(110.2179, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(88.2200, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(425.9008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(150.2174, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(220.3636, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(209.6420, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(75.4944, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(139.5865, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.2206, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(248.9106, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(127.8182, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.9358, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.0505, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4019138755980861, 'recall': 0.5957446808510638, 'f1': 0.48, 'number': 282}, 'P': {'precision': 0.5760869565217391, 'recall': 0.20306513409961685, 'f1': 0.30028328611898014, 'number': 261}, 'overall_precision': 0.43333333333333335, 'overall_recall': 0.40699815837937386, 'overall_f1': 0.4197530864197531, 'overall_accuracy': 0.6245449207281268}
			------------EPOCH 12---------------
Loss:  tensor(149.2676, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(637.4258, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(491.4905, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(522.5596, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(73.0337, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(346.9001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(249.6705, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(285.6563, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.2348, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(91.4266, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(103.1525, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.4248, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(138.9426, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(140.6051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(89.8298, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.5547, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(89.1063, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(149.4708, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.9001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.6323, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(99.1354, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.5213, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(239.0842, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(146.1047, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(328.7798, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(99.6974, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(194.3485, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(297.4355, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(77.6269, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.5738, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(76.8946, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.0443, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(76.1333, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(366.0814, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(194.0394, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.3816, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.5982, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(98.1047, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(75.0146, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(95.3061, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(158.9388, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.3996, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(121.3156, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.8897, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.5030, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(66.6909, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(76.6371, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(73.2429, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(117.6004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(84.9838, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(115.9280, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.8165, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(88.5324, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(184.0471, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.6405, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(67.1828, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5092936802973977, 'recall': 0.4858156028368794, 'f1': 0.49727767695099817, 'number': 282}, 'P': {'precision': 0.4472573839662447, 'recall': 0.4061302681992337, 'f1': 0.4257028112449799, 'number': 261}, 'overall_precision': 0.48023715415019763, 'overall_recall': 0.44751381215469616, 'overall_f1': 0.46329837940896096, 'overall_accuracy': 0.6422196124486201}
			------------EPOCH 13---------------
Loss:  tensor(104.4269, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(176.5757, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(162.9888, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(161.5480, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.5299, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(90.1587, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.2635, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(245.4919, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(157.9762, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(186.5117, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(234.6231, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(148.8081, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(160.5692, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(257.9229, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(328.5668, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(272.3877, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(210.6883, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(233.3488, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(104.9260, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(278.1074, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(67.3705, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(97.5128, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(90.6728, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(72.4733, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(145.3572, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(76.0706, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(104.8668, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(90.5577, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(64.4245, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(70.3944, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.4599, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.5545, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.0386, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(177.0726, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(170.8723, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.2976, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.0881, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(370.6841, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(208.4471, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(284.6570, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(217.9091, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(64.8457, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(565.3622, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.3446, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(78.9689, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(303.3618, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(242.4024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(243.2122, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(232.1913, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(127.0762, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(118.4940, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(95.6686, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(98.4546, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(108.7005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.1190, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(71.6312, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4166666666666667, 'recall': 0.33687943262411346, 'f1': 0.37254901960784315, 'number': 282}, 'P': {'precision': 0.45918367346938777, 'recall': 0.5172413793103449, 'f1': 0.4864864864864865, 'number': 261}, 'overall_precision': 0.44061302681992337, 'overall_recall': 0.42357274401473294, 'overall_f1': 0.431924882629108, 'overall_accuracy': 0.6435114503816793}
			------------EPOCH 14---------------
Loss:  tensor(33.4038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(170.6837, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(137.7656, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(116.1517, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(109.3554, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.4931, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.9013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(339.1071, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(125.3995, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(92.4363, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(161.8589, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(110.7108, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(128.9825, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(141.6601, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(215.3937, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(69.7990, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(112.0511, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(234.1275, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.5956, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(256.3547, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(315.7523, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(266.2609, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(205.3956, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(132.6584, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(156.3599, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(125.7515, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(376.7245, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(353.7473, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.7244, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.2812, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.2973, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.6319, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.8561, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(72.0805, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(78.6211, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.7970, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.8238, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(110.3810, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.5589, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.0646, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(182.6571, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.2190, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(86.3905, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.9514, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(141.6120, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(189.5947, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(102.0149, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(164.7784, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(565.6467, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(316.1458, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(433.6072, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(145.7597, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(439.8226, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(378.7237, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(109.8778, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(232.6729, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.49466192170818507, 'recall': 0.4929078014184397, 'f1': 0.49378330373001783, 'number': 282}, 'P': {'precision': 0.4756756756756757, 'recall': 0.6743295019157088, 'f1': 0.5578446909667196, 'number': 261}, 'overall_precision': 0.4838709677419355, 'overall_recall': 0.580110497237569, 'overall_f1': 0.5276381909547739, 'overall_accuracy': 0.6428655314151497}
			------------EPOCH 15---------------
Loss:  tensor(63.1645, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(168.9594, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(285.8884, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(339.6132, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.3480, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(135.8440, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.1929, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(158.4756, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.3388, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(85.5569, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(128.3018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(66.4198, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(91.2776, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(124.0762, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(91.3554, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(65.8079, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(75.6082, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.0085, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.3916, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.9885, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(149.0182, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(97.4527, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(144.1401, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(106.9391, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(249.8179, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(81.3085, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(224.2005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(170.7206, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(183.6065, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(537.7044, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(152.2447, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(222.3618, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(116.8693, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(162.9166, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(463.5267, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(131.9716, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(95.5185, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(129.8654, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(131.1826, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(142.1410, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(144.3889, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.4539, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(141.7533, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.5787, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.8756, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.4683, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(62.9838, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(70.8225, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.8077, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.2424, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.4663, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.6862, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(71.7934, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(89.6696, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.8217, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.5603, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.38144329896907214, 'recall': 0.13120567375886524, 'f1': 0.19525065963060684, 'number': 282}, 'P': {'precision': 0.3789279112754159, 'recall': 0.7854406130268199, 'f1': 0.511221945137157, 'number': 261}, 'overall_precision': 0.3793103448275862, 'overall_recall': 0.44567219152854515, 'overall_f1': 0.409822184589331, 'overall_accuracy': 0.5623018203170875}
			------------EPOCH 16---------------
Loss:  tensor(425.2571, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(779.8442, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(609.5933, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(759.2178, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(303.1021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(437.3376, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(308.7200, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(257.5016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(65.7134, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(74.3749, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(141.2064, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.5163, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(77.2636, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(117.2682, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(87.4806, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(61.9759, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(64.1934, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(108.9607, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(66.6959, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.5675, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(246.3498, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(154.0419, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(142.0696, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(114.0946, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(339.3060, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(145.4523, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(451.7723, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(293.8853, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(97.0453, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(161.1596, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(59.2171, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(162.5764, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(118.1173, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(104.8348, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(328.4034, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(265.4651, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(333.8960, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(781.8512, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(162.0891, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(166.4220, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(240.9583, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.0912, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(199.9735, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(65.1800, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.7615, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.6620, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(95.9768, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(100.1538, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.1921, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.1137, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(88.7615, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.5446, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(124.3449, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(129.4972, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.6272, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(117.6635, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4418604651162791, 'recall': 0.1347517730496454, 'f1': 0.20652173913043478, 'number': 282}, 'P': {'precision': 0.39956331877729256, 'recall': 0.7011494252873564, 'f1': 0.5090403337969402, 'number': 261}, 'overall_precision': 0.40625, 'overall_recall': 0.40699815837937386, 'overall_f1': 0.406623735050598, 'overall_accuracy': 0.6159718144450969}
			------------EPOCH 17---------------
Loss:  tensor(85.1160, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(278.1036, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(295.6322, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(308.4524, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(132.3116, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(232.4983, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(254.0105, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(435.4424, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1165.7192, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(683.2262, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(432.0208, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(253.2342, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(597.2330, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(449.8268, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(701.0400, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(505.9447, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(66.8638, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(159.4927, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(61.4767, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(93.2853, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.5604, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.8700, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.6268, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.8066, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(62.2284, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.7542, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.1907, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.2323, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.6792, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.2032, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.2630, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.3106, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.4048, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(74.1840, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(73.9366, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.1493, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(108.9941, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(201.9764, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(59.0226, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(187.5345, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(283.5283, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(128.6447, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(302.4158, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(184.2180, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(248.2056, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(150.1302, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(523.0433, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(230.9238, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(282.1529, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(77.0549, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(237.2152, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(206.7701, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(234.3069, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(320.6313, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.7890, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(107.7856, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4945945945945946, 'recall': 0.648936170212766, 'f1': 0.5613496932515337, 'number': 282}, 'P': {'precision': 0.5257352941176471, 'recall': 0.5478927203065134, 'f1': 0.5365853658536586, 'number': 261}, 'overall_precision': 0.5077881619937694, 'overall_recall': 0.6003683241252302, 'overall_f1': 0.5502109704641349, 'overall_accuracy': 0.6489136817381093}
			------------EPOCH 18---------------
Loss:  tensor(61.1982, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.4723, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(76.1156, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(64.4590, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.6950, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.6821, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.6368, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.7994, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.5285, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.5677, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.2364, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.4606, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(74.7947, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(222.4519, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(106.9046, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.3058, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.5711, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.2839, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.6150, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.0474, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.3117, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(82.4888, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(78.5520, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(117.6190, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(144.1821, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.4050, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(112.8871, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(200.8956, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(82.0058, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(133.2716, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.1048, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.6855, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.4090, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.0695, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(95.5970, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.0656, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.7240, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.1732, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.8780, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.1380, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(94.4044, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.0682, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.8179, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.7887, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.0100, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.7936, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.4308, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.1796, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.9438, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.2058, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.3813, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.4493, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.2655, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.1093, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.8545, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.2323, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.48484848484848486, 'recall': 0.45390070921985815, 'f1': 0.46886446886446886, 'number': 282}, 'P': {'precision': 0.4609375, 'recall': 0.6781609195402298, 'f1': 0.5488372093023256, 'number': 261}, 'overall_precision': 0.470679012345679, 'overall_recall': 0.5616942909760589, 'overall_f1': 0.512174643157011, 'overall_accuracy': 0.6344098649442161}
			------------EPOCH 19---------------
Loss:  tensor(9.5020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.9140, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.8530, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.4648, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.6800, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.9776, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.7294, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.8034, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.2510, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.6285, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.2509, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.0596, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.8629, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(82.2043, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.9870, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.0781, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.1669, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.4335, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.8649, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.8433, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.5141, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.1256, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.7851, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.9028, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.8094, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.3669, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.3617, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.8549, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.6948, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.8635, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.1748, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.7764, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.5147, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.2842, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.8737, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.1574, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.9541, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.2849, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.2679, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.3862, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.6121, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.1054, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.9209, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.1339, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.5893, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.0896, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.3650, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.7946, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.2389, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.3928, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.1475, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.5307, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.0803, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.1289, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.9700, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.2742, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5094339622641509, 'recall': 0.574468085106383, 'f1': 0.54, 'number': 282}, 'P': {'precision': 0.4937888198757764, 'recall': 0.6091954022988506, 'f1': 0.5454545454545455, 'number': 261}, 'overall_precision': 0.5015625, 'overall_recall': 0.5911602209944752, 'overall_f1': 0.5426880811496196, 'overall_accuracy': 0.6685848502642395}
			------------EPOCH 20---------------
Loss:  tensor(7.6034, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.1441, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.6395, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.3444, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.0547, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.3358, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.0401, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.7231, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.6472, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.6497, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.7287, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.0417, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.5911, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(73.2429, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.3458, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.9756, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.3491, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.3251, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.6183, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.0303, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.3418, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.1494, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.1747, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.5569, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.6221, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.4573, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.9340, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.2733, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.1210, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.0532, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.2727, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.2614, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.6720, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.7513, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.7515, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.3958, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.0989, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.7775, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.8690, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.7679, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.3986, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.3972, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.0608, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.4905, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.1690, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.4241, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.6789, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.9063, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.8717, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.6792, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.9052, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.8715, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.6611, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.3921, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.2936, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.5994, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5144694533762058, 'recall': 0.5673758865248227, 'f1': 0.5396290050590219, 'number': 282}, 'P': {'precision': 0.5225806451612903, 'recall': 0.6206896551724138, 'f1': 0.5674255691768827, 'number': 261}, 'overall_precision': 0.5185185185185185, 'overall_recall': 0.5930018416206262, 'overall_f1': 0.5532646048109965, 'overall_accuracy': 0.6774515560775103}
			------------EPOCH 21---------------
Loss:  tensor(5.9399, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.3912, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.8000, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.5286, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.4364, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.7503, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.0984, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.9374, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.8979, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.1222, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.2056, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.5231, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.6975, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(71.7720, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.9295, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.1791, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.5395, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.6275, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.5186, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.8719, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.5468, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.2134, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.2372, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.3336, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.5349, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.6642, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.8190, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.3093, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.3425, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.5297, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.5300, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.1078, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.8375, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.5634, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.9261, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.0412, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.9384, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.7875, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.4771, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.4770, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.7784, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.4845, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.1331, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.3118, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.0554, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.6848, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.9019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.2162, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.4276, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.5768, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.8905, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.7355, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.9163, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.3092, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.1662, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.0598, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5098039215686274, 'recall': 0.5531914893617021, 'f1': 0.5306122448979591, 'number': 282}, 'P': {'precision': 0.5125786163522013, 'recall': 0.6245210727969349, 'f1': 0.5630397236614854, 'number': 261}, 'overall_precision': 0.5112179487179487, 'overall_recall': 0.5874769797421732, 'overall_f1': 0.5467009425878321, 'overall_accuracy': 0.6740458015267176}
			------------EPOCH 22---------------
Loss:  tensor(4.7115, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.4073, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.5627, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.5703, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.6254, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.3120, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.1106, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.9310, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.6689, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.0599, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.7074, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.8875, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.9054, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(71.8930, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.4813, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.0956, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.2168, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.8556, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.1698, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.1329, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.0764, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.7944, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.8851, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.7429, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.8214, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.1842, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.4476, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.2424, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.0572, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.2213, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.8650, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.7151, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.6654, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.9336, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.9046, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.9914, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.8993, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.2015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.7932, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.7277, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.9845, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.7189, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.4686, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.4758, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.0483, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.5754, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.1451, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.4945, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.4461, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.2264, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.0127, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.8911, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.4437, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.1778, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.1818, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.8669, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5098039215686274, 'recall': 0.5531914893617021, 'f1': 0.5306122448979591, 'number': 282}, 'P': {'precision': 0.5, 'recall': 0.6283524904214559, 'f1': 0.5568760611205433, 'number': 261}, 'overall_precision': 0.5047318611987381, 'overall_recall': 0.5893186003683242, 'overall_f1': 0.5437553101104503, 'overall_accuracy': 0.6726952436876101}
			------------EPOCH 23---------------
Loss:  tensor(3.7867, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.2279, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.1610, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.3445, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.2726, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.2358, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.4072, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.7127, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.9050, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.2010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.8358, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.5933, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.9211, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(76.8746, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.0500, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.7060, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.4836, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.3588, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.6857, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.6420, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.0659, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.6936, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.8090, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.1376, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.7404, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.9183, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.6968, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.4761, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.2721, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.2224, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.7643, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.9101, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.7170, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.6876, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.2857, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.1587, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.0494, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.8687, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.3618, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.3247, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.3731, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.5350, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.7671, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.8589, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.1952, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.1967, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.7541, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.2335, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.7260, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.5438, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.5808, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.2905, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.2661, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.9488, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.3317, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.8428, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5098039215686274, 'recall': 0.5531914893617021, 'f1': 0.5306122448979591, 'number': 282}, 'P': {'precision': 0.5015290519877675, 'recall': 0.6283524904214559, 'f1': 0.5578231292517006, 'number': 261}, 'overall_precision': 0.5055292259083728, 'overall_recall': 0.5893186003683242, 'overall_f1': 0.5442176870748299, 'overall_accuracy': 0.6727539635936582}
			------------EPOCH 24---------------
Loss:  tensor(3.2289, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.1413, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.7896, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.5482, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.1045, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.6636, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.0169, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.1629, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.4754, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.5934, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.6926, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.9254, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.2246, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.7566, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.3689, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.2106, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.2052, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.3064, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.6715, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.2679, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.2931, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.8505, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.0589, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.3811, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.1622, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.0589, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.8139, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.9021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.9874, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.4563, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.9529, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.4835, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.9732, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.6291, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.0899, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.5452, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.4987, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.7504, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.1568, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.2788, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.8521, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.8971, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.5602, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.1165, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.5511, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.1822, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.7724, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.3168, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.1770, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.1128, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.5533, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.8461, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.4557, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.4544, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.7172, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.1286, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5261437908496732, 'recall': 0.5709219858156028, 'f1': 0.5476190476190477, 'number': 282}, 'P': {'precision': 0.49848024316109424, 'recall': 0.6283524904214559, 'f1': 0.5559322033898306, 'number': 261}, 'overall_precision': 0.5118110236220472, 'overall_recall': 0.5985267034990792, 'overall_f1': 0.5517826825127335, 'overall_accuracy': 0.6729301233118027}
			------------EPOCH 25---------------
Loss:  tensor(2.7998, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.8518, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.3239, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.9881, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.1977, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.0567, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.7485, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.7186, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.6841, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.3932, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.8164, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.2349, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.9577, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.8394, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.7391, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.0988, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.3884, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.4779, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.9106, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.9074, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.6399, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.1434, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.4281, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.3942, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.9699, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.9166, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.7818, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.6615, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.4384, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.8685, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.3212, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.1622, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.3622, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.8161, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.1833, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.0913, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.9743, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.2754, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.1631, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.4473, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.0359, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.6638, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.5838, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.2492, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.1454, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.4375, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.9909, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.5608, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.7329, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.7769, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.6790, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5186, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.4617, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.0300, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.2173, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.4368, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5309446254071661, 'recall': 0.5780141843971631, 'f1': 0.5534804753820034, 'number': 282}, 'P': {'precision': 0.5, 'recall': 0.6283524904214559, 'f1': 0.5568760611205433, 'number': 261}, 'overall_precision': 0.5149606299212598, 'overall_recall': 0.6022099447513812, 'overall_f1': 0.5551782682512734, 'overall_accuracy': 0.6752201996476805}
			------------EPOCH 26---------------
Loss:  tensor(2.4825, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.7266, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.3028, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.5150, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.4403, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.8991, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.3322, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.8126, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.9962, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.3163, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.2120, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.8425, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.6383, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.9412, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.3373, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.9802, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.7079, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.5337, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.3421, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.6353, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.1723, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.6664, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.9559, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.2013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.2277, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.9771, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.3420, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.3532, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.8510, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.4412, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.7793, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.9631, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.8518, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.9944, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.4503, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.7276, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.5382, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.4711, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.3054, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.7055, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.2549, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.6592, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.7993, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.4193, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.8522, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.8577, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.3841, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.8721, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.3655, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5064, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.9364, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2354, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.1129, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.4028, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.8387, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.9141, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5309446254071661, 'recall': 0.5780141843971631, 'f1': 0.5534804753820034, 'number': 282}, 'P': {'precision': 0.4940119760479042, 'recall': 0.632183908045977, 'f1': 0.5546218487394958, 'number': 261}, 'overall_precision': 0.5117004680187207, 'overall_recall': 0.6040515653775322, 'overall_f1': 0.5540540540540541, 'overall_accuracy': 0.6757486788021139}
			------------EPOCH 27---------------
Loss:  tensor(2.2020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.8089, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.3012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.1575, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.7509, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.5747, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.8265, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.3582, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.4982, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.3789, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.5895, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.5351, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.5783, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.2437, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.8917, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.1918, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.4932, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.0824, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.8672, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.3717, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.7034, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.2461, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.5323, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.2135, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.1753, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.6932, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.2140, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.4382, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.7252, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.0423, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.3545, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.8021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.4295, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.4273, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.8784, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.4128, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.2251, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.7633, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.4687, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.0955, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.0646, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.8067, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.0754, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.1851, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.5326, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.3988, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.9938, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.2701, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.0501, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2697, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.3715, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.9956, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.9369, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.0564, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5391, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.5391, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5276872964169381, 'recall': 0.574468085106383, 'f1': 0.5500848896434635, 'number': 282}, 'P': {'precision': 0.49700598802395207, 'recall': 0.6360153256704981, 'f1': 0.5579831932773108, 'number': 261}, 'overall_precision': 0.5117004680187207, 'overall_recall': 0.6040515653775322, 'overall_f1': 0.5540540540540541, 'overall_accuracy': 0.6758073987081621}
			------------EPOCH 28---------------
Loss:  tensor(1.9698, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.0246, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.4912, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.8628, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.1831, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.4789, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.3837, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.9637, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.0879, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.4974, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.9838, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.5283, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.5536, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.5512, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.5365, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.5617, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.9694, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.3438, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5065, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.1558, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.3402, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.9159, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.1671, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.0663, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.0709, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.5310, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.1205, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.4871, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.1404, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.7255, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.9647, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.7545, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.0267, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.9158, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.3796, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.1511, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.8776, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.1523, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.7929, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.5341, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.6027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.1510, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.4211, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.0826, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.4489, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.0346, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.6125, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.6744, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.7786, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.0710, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.8435, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.7977, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.9111, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.9530, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2830, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.1714, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5276872964169381, 'recall': 0.574468085106383, 'f1': 0.5500848896434635, 'number': 282}, 'P': {'precision': 0.4911242603550296, 'recall': 0.6360153256704981, 'f1': 0.5542570951585977, 'number': 261}, 'overall_precision': 0.5085271317829457, 'overall_recall': 0.6040515653775322, 'overall_f1': 0.5521885521885522, 'overall_accuracy': 0.6739870816206694}
			------------EPOCH 29---------------
Loss:  tensor(1.7737, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.3094, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.7541, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.6063, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.7030, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.5560, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.9004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.5682, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.7470, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.6808, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.5443, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.7787, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.6142, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.1162, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.2274, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.8240, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.8781, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.0126, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2202, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.9658, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.0173, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.6390, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.8596, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.9663, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.0315, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.4782, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.1138, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.6508, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.9392, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.4480, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.6229, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.8699, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.6746, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.4853, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.9557, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.9288, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.6272, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.6154, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.1826, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.0520, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.6469, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.6110, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.8620, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.0209, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.5126, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.7191, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.3875, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.1244, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.5231, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.8807, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.4179, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6232, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.9894, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.9822, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.0544, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.8803, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5313531353135313, 'recall': 0.5709219858156028, 'f1': 0.5504273504273504, 'number': 282}, 'P': {'precision': 0.4956268221574344, 'recall': 0.6513409961685823, 'f1': 0.5629139072847682, 'number': 261}, 'overall_precision': 0.5123839009287926, 'overall_recall': 0.6095764272559853, 'overall_f1': 0.5567703952901598, 'overall_accuracy': 0.6745155607751028}
			------------EPOCH 30---------------
Loss:  tensor(1.6051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.6884, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.1332, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.3794, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.2403, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.6526, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.4947, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.2256, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.4159, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.8916, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.1841, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.2238, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.7288, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.6560, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.9068, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.3548, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.6129, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.5382, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.9757, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.7957, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.7181, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.3896, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.5869, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.8930, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.1024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.5170, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.1873, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.9010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.9337, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2054, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.3138, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.1454, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.3507, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.1195, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.6067, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.7229, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.3749, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.1266, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.6151, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.6234, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.9397, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.1486, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.3553, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.5097, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.7546, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.4270, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.1905, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.5773, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.2911, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.7126, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.0260, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4751, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.1682, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.1463, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.8561, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.6097, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5331125827814569, 'recall': 0.5709219858156028, 'f1': 0.5513698630136986, 'number': 282}, 'P': {'precision': 0.4941860465116279, 'recall': 0.6513409961685823, 'f1': 0.5619834710743802, 'number': 261}, 'overall_precision': 0.5123839009287926, 'overall_recall': 0.6095764272559853, 'overall_f1': 0.5567703952901598, 'overall_accuracy': 0.6745742806811509}


		-------------RUN 2-----------
			------------EPOCH 1---------------
Loss:  tensor(1960.6418, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2831.3430, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2569.1953, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3109.0547, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2042.8668, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1939.4075, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1856.8743, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2292.8384, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4218.0610, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3457.4133, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3194.7666, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2734.2603, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2241.3992, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2596.5669, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1631.9012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1883.1562, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2439.5645, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2022.9609, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2144.4519, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2507.0642, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1167.2245, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(792.1301, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1007.2995, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1133.4901, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2093.8423, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1980.1244, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1310.8104, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1506.2776, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1195.0120, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1902.9304, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1327.6816, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2175.2195, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1183.4497, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1274.1854, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1348.6187, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1988.1345, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(946.0099, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1657.3602, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1597.8933, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1712.9427, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2034.4417, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2003.8180, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(996.9041, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1396.8750, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1606.8090, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(425.4064, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(991.8531, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1412.3495, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1470.9161, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1740.1680, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2759.1001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1768.6224, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1989.9045, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2583.2634, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2270.9534, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1614.3850, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1640.2460, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(773.5658, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1386.2921, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.08547008547008547, 'recall': 0.04184100418410042, 'f1': 0.056179775280898875, 'number': 239}, 'P': {'precision': 0.27835051546391754, 'recall': 0.15, 'f1': 0.19494584837545123, 'number': 360}, 'overall_precision': 0.2057877813504823, 'overall_recall': 0.10684474123539232, 'overall_f1': 0.14065934065934063, 'overall_accuracy': 0.5198861508501236}
			------------EPOCH 2---------------
Loss:  tensor(1242.6465, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1765.5261, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1477.7697, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1855.3638, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1409.9806, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1270.5050, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1237.8450, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1458.6757, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3032.5840, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2403.0293, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2222.1353, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2052.4995, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1677.7504, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1894.0562, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1179.4856, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1357.4076, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1856.4526, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1432.5674, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1678.6678, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1931.5371, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(927.8553, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(606.1808, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(829.6840, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(864.6182, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1525.0077, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1568.7551, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1010.4557, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1082.1990, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(894.6124, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1443.5309, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1105.1440, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1884.8539, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(952.3145, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(998.2314, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1017.9501, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1522.6345, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(802.5229, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1373.8812, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1334.3419, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1248.7551, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1667.4911, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1739.2788, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(828.4967, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1190.9103, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1433.7349, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(359.9997, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(866.1914, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1227.9559, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1215.9553, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1373.7056, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2383.2712, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1229.3990, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1562.5760, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2158.0784, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1904.8842, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1218.4377, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1254.8354, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(658.2503, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1161.7854, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.25, 'recall': 0.07531380753138076, 'f1': 0.11575562700964631, 'number': 239}, 'P': {'precision': 0.5011286681715575, 'recall': 0.6166666666666667, 'f1': 0.5529265255292652, 'number': 360}, 'overall_precision': 0.46601941747572817, 'overall_recall': 0.4006677796327212, 'overall_f1': 0.4308797127468582, 'overall_accuracy': 0.6601003670137069}
			------------EPOCH 3---------------
Loss:  tensor(960.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1374.5759, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(987.1531, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1491.0438, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1221.7887, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1083.8451, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1113.4399, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1264.3818, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2692.1704, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2090.9661, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1829.0776, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1827.0684, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1409.1449, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1555.9907, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(988.7217, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1154.6035, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1593.4592, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1264.3757, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1286.0406, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1590.5530, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(813.8085, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(533.2365, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(705.2770, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(727.4863, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1464.9629, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1426.6960, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(937.4849, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(825.3927, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(737.3350, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1172.2008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(955.5553, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1542.3701, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(823.3290, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(817.4967, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(842.0237, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1264.6974, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(704.0851, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1186.0914, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1014.9410, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1058.2805, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1464.6410, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1238.5005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(626.4175, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(798.5243, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1126.0911, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(308.7946, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(685.7317, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1094.4773, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1087.6990, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(998.1165, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1708.3483, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(988.0870, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1313.9299, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1709.1168, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1610.7438, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1076.6088, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1089.9032, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(559.6071, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(927.7396, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.36363636363636365, 'recall': 0.100418410041841, 'f1': 0.15737704918032785, 'number': 239}, 'P': {'precision': 0.4387755102040816, 'recall': 0.35833333333333334, 'f1': 0.39449541284403666, 'number': 360}, 'overall_precision': 0.425, 'overall_recall': 0.25542570951585974, 'overall_f1': 0.31908237747653806, 'overall_accuracy': 0.5831023893341323}
			------------EPOCH 4---------------
Loss:  tensor(903.0502, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1083.0879, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(765.5442, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1333.0308, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1031.7660, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(912.7245, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1038.6005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1313.4890, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2318.7354, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1688.8763, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1559.7286, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1596.7712, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1359.3959, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1279.4032, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(792.2166, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(847.5507, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1483.1714, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1170.6072, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1133.9618, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1495.0276, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(649.1495, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(429.4370, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(581.7355, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(671.4845, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(963.4911, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1160.7694, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(684.9167, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(689.5246, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(600.6365, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1025.6477, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(853.2212, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1349.6177, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(685.6646, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(745.4577, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(759.8152, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1159.9048, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(580.0991, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(868.0669, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(751.6691, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(865.9828, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1202.7794, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(802.7596, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(459.1472, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(446.8544, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(809.6844, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(226.9657, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(548.1072, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(949.6122, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(754.2935, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(789.8311, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1368.9424, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(882.8303, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(953.2924, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1337.5940, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1244.6001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(884.2686, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(809.5223, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(379.7097, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(629.7120, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3372093023255814, 'recall': 0.24267782426778242, 'f1': 0.2822384428223844, 'number': 239}, 'P': {'precision': 0.5551948051948052, 'recall': 0.475, 'f1': 0.5119760479041916, 'number': 360}, 'overall_precision': 0.47708333333333336, 'overall_recall': 0.3823038397328882, 'overall_f1': 0.42446709916589437, 'overall_accuracy': 0.674031907722268}
			------------EPOCH 5---------------
Loss:  tensor(695.4508, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(874.5616, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(569.7920, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1072.4790, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(789.3524, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(678.8657, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(764.0292, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(811.6053, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2032.4010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1455.0979, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1198.0394, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1294.3690, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(988.2971, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(976.0912, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(618.8401, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(659.0858, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1187.8196, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(884.2283, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(793.4844, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1152.4377, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(510.8108, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(345.0298, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(466.3399, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(574.5399, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(646.6036, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(899.5430, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(682.1859, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(510.8938, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(430.5919, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(761.1887, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(742.0677, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1211.1833, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(471.8840, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(545.7476, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(557.4567, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(844.9186, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(555.6512, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(665.3394, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(571.2146, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(765.0450, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1232.3080, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(799.3523, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(420.4855, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(402.5185, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(664.7206, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(164.8341, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(485.2200, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(664.0891, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(573.1195, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(531.0712, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(968.7635, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(701.9868, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(838.6194, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1160.2990, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1140.2986, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(737.9331, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(685.4394, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(345.8849, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(592.3400, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.41735537190082644, 'recall': 0.4225941422594142, 'f1': 0.4199584199584199, 'number': 239}, 'P': {'precision': 0.6619718309859155, 'recall': 0.6527777777777778, 'f1': 0.6573426573426574, 'number': 360}, 'overall_precision': 0.5628140703517588, 'overall_recall': 0.5609348914858097, 'overall_f1': 0.5618729096989967, 'overall_accuracy': 0.7510298854018426}
			------------EPOCH 6---------------
Loss:  tensor(570.3463, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(675.5663, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(422.6299, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(863.9058, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(607.9362, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(524.5767, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(569.6856, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(510.9806, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1484.9535, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1034.0320, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(941.1633, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(884.6820, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(734.9877, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(847.2506, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(549.0579, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(495.4338, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(789.7313, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(606.3462, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(646.5619, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(784.3989, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(331.1900, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(215.3958, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(314.1161, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(407.9430, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(398.7082, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(474.4861, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(692.6586, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(429.6681, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(323.1183, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(590.2847, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(624.7611, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1014.8272, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(399.2486, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(487.3820, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(482.8364, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(736.8442, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(400.0815, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(692.1563, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(365.5666, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(550.6080, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(741.6795, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(478.3205, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(282.2314, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(159.1200, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(483.6967, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(207.3710, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(407.8334, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(679.9299, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(700.1647, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(476.4929, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(920.9740, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(642.2032, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(506.4301, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(958.2999, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(856.2012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(602.0361, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(660.0308, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(251.7228, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(461.6464, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4461538461538462, 'recall': 0.24267782426778242, 'f1': 0.3143631436314363, 'number': 239}, 'P': {'precision': 0.592814371257485, 'recall': 0.825, 'f1': 0.6898954703832751, 'number': 360}, 'overall_precision': 0.5625990491283677, 'overall_recall': 0.5926544240400667, 'overall_f1': 0.5772357723577236, 'overall_accuracy': 0.736349337128305}
			------------EPOCH 7---------------
Loss:  tensor(607.7514, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(608.9871, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(339.8932, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(872.4130, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(546.1663, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(517.2455, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(652.3345, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(558.9995, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1358.1333, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(835.0344, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(863.8309, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(793.9464, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(740.3379, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(686.3846, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(472.9585, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(346.6248, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(905.9338, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(745.7173, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(456.1645, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(790.5390, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(244.8616, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(211.6663, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(281.3495, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(363.2375, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(636.9633, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(463.8653, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(533.6375, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(311.1197, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(242.9101, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(513.8409, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(413.3653, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(775.4305, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(422.8312, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(493.8541, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(625.7874, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(803.8722, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(365.2488, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(482.9626, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(241.6342, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(521.1967, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(664.6776, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(461.5875, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(351.9827, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(173.8775, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(430.3802, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(149.8593, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(376.4754, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(510.4084, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(444.8164, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(458.0780, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(775.3677, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(562.8551, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(429.5329, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1077.1437, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(693.4530, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(715.5988, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(557.1349, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(231.8889, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(370.3557, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4406779661016949, 'recall': 0.4351464435146444, 'f1': 0.4378947368421053, 'number': 239}, 'P': {'precision': 0.6192052980132451, 'recall': 0.5194444444444445, 'f1': 0.5649546827794562, 'number': 360}, 'overall_precision': 0.5408921933085502, 'overall_recall': 0.48580968280467446, 'overall_f1': 0.5118733509234827, 'overall_accuracy': 0.7360497341023144}
			------------EPOCH 8---------------
Loss:  tensor(460.8348, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(438.9999, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(230.1591, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(453.9185, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(318.9764, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(422.2375, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(497.1599, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(347.5053, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1524.2971, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1057.9041, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(801.9243, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(787.9033, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(511.2632, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(782.4708, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(484.5172, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(364.6878, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(500.4302, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(457.1494, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(361.4050, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(437.9764, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(155.6553, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(131.2931, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(197.3523, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(263.6224, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(330.2613, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(315.3574, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(321.8168, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(238.2337, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(330.0611, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(403.0333, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(580.0579, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(627.6672, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(369.4976, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(408.1116, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(326.6788, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(528.2018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(278.2744, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(391.1680, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(268.0829, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(324.5059, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(457.4612, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(282.7914, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(183.6567, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(95.8858, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(275.4536, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(106.9945, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(252.5673, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(307.3495, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(324.7637, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(241.8121, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(446.2908, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(380.7090, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(304.7629, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(574.2623, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(474.3405, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(374.3553, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(324.5360, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(138.5853, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(247.3207, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4322033898305085, 'recall': 0.42677824267782427, 'f1': 0.4294736842105263, 'number': 239}, 'P': {'precision': 0.6580645161290323, 'recall': 0.5666666666666667, 'f1': 0.6089552238805971, 'number': 360}, 'overall_precision': 0.5604395604395604, 'overall_recall': 0.5108514190317195, 'overall_f1': 0.5344978165938865, 'overall_accuracy': 0.7305819788779867}
			------------EPOCH 9---------------
Loss:  tensor(303.7957, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(221.2661, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(136.2124, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(277.6675, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(212.7226, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(268.1658, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(251.7971, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(191.8933, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(713.1790, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(489.6285, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(414.2236, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(284.7126, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(292.0662, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(509.9311, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(277.6173, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(145.7068, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(303.4837, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(341.2035, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(210.2450, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(250.3946, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(137.4892, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(81.2632, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(99.9953, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(205.1362, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(126.4661, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(126.8402, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(305.3465, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(143.3284, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(120.8547, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(249.5695, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(241.2896, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(426.1427, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(215.0688, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(307.8333, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(246.7359, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(330.5583, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(157.1612, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(135.5056, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(104.1953, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(143.2985, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(366.7373, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(213.5849, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(108.7065, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(69.1591, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(218.6602, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(106.9262, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(210.1999, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(226.2893, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(233.4392, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(217.3526, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(446.8110, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(349.9550, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(174.6344, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(331.4833, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(196.3842, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(301.2504, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(191.9901, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(101.7080, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(199.0315, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.46534653465346537, 'recall': 0.39330543933054396, 'f1': 0.4263038548752835, 'number': 239}, 'P': {'precision': 0.6632911392405063, 'recall': 0.7277777777777777, 'f1': 0.6940397350993377, 'number': 360}, 'overall_precision': 0.5963149078726968, 'overall_recall': 0.5943238731218697, 'overall_f1': 0.5953177257525083, 'overall_accuracy': 0.7670586472923376}
			------------EPOCH 10---------------
Loss:  tensor(215.5237, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(112.8769, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(76.7110, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(151.3427, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(108.2128, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(252.9192, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(170.0463, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(154.9623, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(598.7232, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(418.5630, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(330.4164, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(179.6362, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(180.7653, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(449.1426, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(177.3487, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(86.8507, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(124.7312, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(200.2027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(160.4381, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(156.2058, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(100.4652, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(80.9953, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(100.8141, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(135.3124, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(138.3096, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(133.6656, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(166.2118, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(87.3727, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(127.8644, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(203.7558, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(277.0171, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(285.2219, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(179.5593, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(234.7257, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(174.0258, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(233.3503, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(92.0583, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(77.4562, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.3705, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(88.2053, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(262.3025, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(128.8314, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(85.7626, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(62.2658, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(159.3346, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.0396, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(164.2963, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(163.4333, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(182.5668, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(185.2259, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(233.0889, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(218.2086, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(179.8469, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(208.0641, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(155.5260, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(257.2183, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(167.7361, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(71.6219, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(173.6438, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3688118811881188, 'recall': 0.6234309623430963, 'f1': 0.463452566096423, 'number': 239}, 'P': {'precision': 0.7610619469026548, 'recall': 0.4777777777777778, 'f1': 0.5870307167235495, 'number': 360}, 'overall_precision': 0.5095238095238095, 'overall_recall': 0.5358931552587646, 'overall_f1': 0.5223759153783563, 'overall_accuracy': 0.7054153246947794}
			------------EPOCH 11---------------
Loss:  tensor(203.6400, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(168.4153, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(69.3263, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(132.2692, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(99.5059, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(113.4930, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(85.5599, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(102.6352, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(270.0729, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(194.3008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(132.8567, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(95.4977, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(120.2816, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(295.8759, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(143.5636, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(69.9218, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(97.1053, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(307.5055, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(197.6447, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(134.4977, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(130.7421, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.8260, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(61.5389, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(188.0388, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(68.0188, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(75.2490, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(162.2527, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(91.2988, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.5419, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(132.2348, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(110.5432, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(221.9006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(147.6508, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(227.8441, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(144.3720, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(186.1703, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(62.5558, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(82.7887, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(77.6121, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(93.0815, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(312.8528, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(118.8839, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(100.2430, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.1326, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(167.9020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(71.8175, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(155.5471, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(123.4598, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(97.1289, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(145.4661, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(286.7050, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(220.9365, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(85.8886, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(111.6048, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(85.7229, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(158.9210, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(76.1078, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.4903, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(115.2784, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.42813455657492355, 'recall': 0.5857740585774058, 'f1': 0.49469964664310956, 'number': 239}, 'P': {'precision': 0.6851851851851852, 'recall': 0.6166666666666667, 'f1': 0.6491228070175439, 'number': 360}, 'overall_precision': 0.5560675883256528, 'overall_recall': 0.6043405676126878, 'overall_f1': 0.5791999999999999, 'overall_accuracy': 0.7305070781214891}
			------------EPOCH 12---------------
Loss:  tensor(72.7674, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.8931, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.5969, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(67.9795, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(66.4637, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(114.7903, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(100.9831, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(68.5523, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(294.7904, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(159.9819, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(176.3901, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(99.1920, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(98.5801, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(247.4630, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(100.0176, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.2246, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(130.0096, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(135.6274, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(126.3254, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(174.3614, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(64.7250, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.0882, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.8247, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(72.3556, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.3861, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(59.0913, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(85.5142, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.2073, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.4610, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(98.3285, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(76.5638, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(160.2658, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(148.5963, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(157.2081, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(219.7433, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(182.4254, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.1444, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.3681, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.8217, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(65.2514, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(209.2260, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(91.3079, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.0315, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.7444, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(156.6023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.4878, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(65.6034, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(148.6422, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(492.1383, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(203.7379, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(602.8129, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(297.7081, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(89.1349, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(363.5936, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(154.3376, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(278.8984, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(77.8351, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.2546, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(85.2866, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4182825484764543, 'recall': 0.6317991631799164, 'f1': 0.5033333333333333, 'number': 239}, 'P': {'precision': 0.725, 'recall': 0.5638888888888889, 'f1': 0.634375, 'number': 360}, 'overall_precision': 0.5522620904836193, 'overall_recall': 0.5909849749582637, 'overall_f1': 0.5709677419354837, 'overall_accuracy': 0.747809152872444}
			------------EPOCH 13---------------
Loss:  tensor(49.5265, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.0332, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.4472, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.6173, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(61.1662, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(64.1657, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.3524, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(59.5555, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(244.1539, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(161.8189, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(104.1384, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(108.6733, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(194.1629, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(322.8646, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(98.2304, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(82.0542, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(359.4276, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(141.5950, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(263.1577, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(297.0650, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(70.4031, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.4294, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(59.8797, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(112.9077, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(131.7313, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(72.9993, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(81.1881, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.0250, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.7484, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(87.5245, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(102.5745, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(218.4373, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(81.1932, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(103.0694, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(146.5052, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(132.6573, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.9304, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(81.5206, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.4731, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(62.9877, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(239.2433, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(108.8768, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(82.8186, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.5479, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(116.9942, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.7161, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(74.0116, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(97.2609, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(115.7508, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(110.8614, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(258.2676, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(173.4194, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(270.6483, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(230.1098, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(137.5230, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(243.2877, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(342.0208, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(152.4784, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(199.2305, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.37721518987341773, 'recall': 0.6234309623430963, 'f1': 0.47003154574132494, 'number': 239}, 'P': {'precision': 0.49107142857142855, 'recall': 0.1527777777777778, 'f1': 0.23305084745762714, 'number': 360}, 'overall_precision': 0.40236686390532544, 'overall_recall': 0.34056761268781305, 'overall_f1': 0.36889692585895123, 'overall_accuracy': 0.596284922477717}
			------------EPOCH 14---------------
Loss:  tensor(552.9604, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(386.9059, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(321.4770, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(411.0460, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(84.4068, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(169.0244, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(75.7051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(241.3337, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(368.5743, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(124.9723, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(89.1860, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.8566, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(75.9514, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(268.5174, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(76.0996, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(72.5576, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(134.0106, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(312.2449, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(91.7295, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(165.1676, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(137.2366, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(72.9809, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(172.7873, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(364.0664, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(111.7206, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(172.4001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(213.0638, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(134.6913, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(109.4924, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(127.9190, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(224.4250, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(300.1122, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(188.0757, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(127.6829, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(179.2407, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(127.7899, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(79.4820, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(163.9613, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(125.0602, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.9943, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(223.2062, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(112.3024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.9799, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(59.5659, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(85.2398, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.8876, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.1934, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(153.7745, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(99.1455, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(108.5109, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(102.7708, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(125.0690, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.2373, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(78.7196, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(73.9987, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(105.0101, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(74.5349, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.9795, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(84.3362, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.38258575197889183, 'recall': 0.606694560669456, 'f1': 0.46925566343042074, 'number': 239}, 'P': {'precision': 0.5987261146496815, 'recall': 0.2611111111111111, 'f1': 0.36363636363636365, 'number': 360}, 'overall_precision': 0.4458955223880597, 'overall_recall': 0.3989983305509182, 'overall_f1': 0.42114537444933925, 'overall_accuracy': 0.6418994831847802}
			------------EPOCH 15---------------
Loss:  tensor(64.0746, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.3170, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.8485, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(83.4846, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(205.7347, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(244.6160, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(191.4015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(148.6647, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1036.3868, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(526.2981, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(410.8983, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(415.5118, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(243.6548, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(272.5428, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(327.4947, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(85.2225, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.7022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.2789, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(87.2785, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(114.1759, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(80.3712, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.8192, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.2357, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(77.2817, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(122.4425, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(157.8844, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(231.4686, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(146.6485, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(450.0016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(552.5886, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1037.2349, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1153.9458, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(968.5347, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(488.9022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(643.3999, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1032.3009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(86.8315, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(69.0567, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(146.2058, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(97.7179, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(379.0798, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(143.9760, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.7865, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.5129, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(259.2496, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(97.5860, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(81.0958, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(147.2535, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(584.9672, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(411.2451, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(855.1443, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1019.7003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(456.1641, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(580.5344, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(383.2554, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(243.9993, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(98.9824, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(78.3554, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(111.9133, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4234875444839858, 'recall': 0.497907949790795, 'f1': 0.4576923076923077, 'number': 239}, 'P': {'precision': 0.5897435897435898, 'recall': 0.25555555555555554, 'f1': 0.3565891472868217, 'number': 360}, 'overall_precision': 0.482837528604119, 'overall_recall': 0.35225375626043404, 'overall_f1': 0.4073359073359073, 'overall_accuracy': 0.6219009811999101}
			------------EPOCH 16---------------
Loss:  tensor(131.2088, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(94.8283, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.3012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(102.6171, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(97.5761, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(195.5784, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(133.5949, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(119.0415, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(648.4757, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(349.3179, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(390.0569, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(156.9293, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(219.6206, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(604.1564, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(381.6516, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(179.9371, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(206.7464, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(169.1649, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(160.5609, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(205.6103, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(71.5828, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.7685, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(86.0425, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(103.3170, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(88.6582, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(72.6524, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(74.8051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(100.6510, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.5160, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(81.9342, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(85.2684, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(76.9516, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(104.4103, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(105.5284, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(95.1687, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(97.0157, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(99.3570, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(141.8059, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.8420, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(142.5603, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(457.6039, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(363.5438, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(205.4001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(114.6757, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(505.6168, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.6219, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(311.0547, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(720.4706, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(382.6844, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(216.8534, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(453.8807, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(514.5676, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(368.4363, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(361.4395, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(251.0874, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(174.6292, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(98.0348, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(70.2745, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(154.4925, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3655913978494624, 'recall': 0.7112970711297071, 'f1': 0.48295454545454536, 'number': 239}, 'P': {'precision': 0.7678571428571429, 'recall': 0.35833333333333334, 'f1': 0.4886363636363637, 'number': 360}, 'overall_precision': 0.47235387045813587, 'overall_recall': 0.4991652754590985, 'overall_f1': 0.4853896103896104, 'overall_accuracy': 0.672159388809827}
			------------EPOCH 17---------------
Loss:  tensor(159.5681, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(266.1954, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(180.2496, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(169.2569, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(323.1259, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(449.2771, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(153.4464, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(215.2256, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(669.0049, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(320.8071, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(311.8638, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(386.9011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(353.6729, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(320.3009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(310.1462, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(179.2688, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(220.0542, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(164.9587, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(78.5072, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(131.7918, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.8010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.4445, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.9665, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(73.5549, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(119.9059, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(132.7769, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(85.3597, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.2806, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(59.9024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(200.5916, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(113.0179, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(277.1551, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(255.6086, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(229.1688, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(201.2611, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(314.7811, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(192.7635, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(265.4146, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(105.8704, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(635.3929, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(673.5284, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(383.2498, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(214.5395, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(211.4846, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(362.7675, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(59.9977, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(137.3116, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(243.7128, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(251.2080, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(194.4417, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(465.8831, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(263.4174, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(161.1714, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(492.7382, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(454.5902, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(251.6868, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(122.7558, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.4932, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.3833, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.43902439024390244, 'recall': 0.602510460251046, 'f1': 0.507936507936508, 'number': 239}, 'P': {'precision': 0.7197231833910035, 'recall': 0.5777777777777777, 'f1': 0.6409861325115561, 'number': 360}, 'overall_precision': 0.5705024311183144, 'overall_recall': 0.5876460767946577, 'overall_f1': 0.5789473684210527, 'overall_accuracy': 0.7487079619504157}
			------------EPOCH 18---------------
Loss:  tensor(48.1661, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.7788, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.7956, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.8600, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.1206, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.5543, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(61.4234, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.9806, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(297.0380, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(124.2407, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(174.7439, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(93.7848, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(266.0658, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(340.1995, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(187.0061, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(129.9480, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1013.2657, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(994.3445, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(410.7181, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(946.4889, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(321.9892, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(274.3712, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(359.9437, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(555.2710, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(222.0662, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(237.7452, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(212.1202, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(205.4982, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(190.3432, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(168.3239, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(269.7460, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(138.7321, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(96.0591, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(85.7246, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(79.1845, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(93.3992, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.1232, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.0379, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.4359, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.0570, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(70.5043, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(64.0975, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.0584, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.8815, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(109.0345, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.3738, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.5906, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(98.3650, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(146.1773, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(135.5337, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(156.7067, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(137.7016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(156.9393, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(551.2415, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(312.2537, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(287.8145, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(479.5171, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(241.2462, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(342.6481, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.6666666666666666, 'recall': 0.26778242677824265, 'f1': 0.3820895522388059, 'number': 239}, 'P': {'precision': 0.6422594142259415, 'recall': 0.8527777777777777, 'f1': 0.7326968973747018, 'number': 360}, 'overall_precision': 0.6463414634146342, 'overall_recall': 0.6193656093489148, 'overall_f1': 0.6325660699062234, 'overall_accuracy': 0.7562729383566774}
			------------EPOCH 19---------------
Loss:  tensor(333.0708, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(272.0991, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(147.2253, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(378.8333, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(95.6992, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(196.5915, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(413.9613, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(376.2377, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(287.1425, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(280.2072, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(105.7786, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(99.7431, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.6161, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(130.8643, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(76.5304, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(79.6465, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.3474, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(59.8862, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(69.6655, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.5904, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.6486, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.8298, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.3577, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.2911, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(70.6267, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(102.8786, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(70.5163, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.2819, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.2518, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(90.1282, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(169.7422, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(150.9308, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(187.9842, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(127.5135, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(130.0488, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(296.4883, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(190.4375, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(472.8725, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(307.3110, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(133.3219, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(274.3310, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(164.2610, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.5561, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.3915, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(116.9430, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.8130, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(74.6474, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(82.6546, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.0272, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(78.0655, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(70.2388, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(81.2785, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.5847, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(73.8703, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.9435, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(68.5446, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.1820, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.6610, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.9979, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4895833333333333, 'recall': 0.5899581589958159, 'f1': 0.5351043643263758, 'number': 239}, 'P': {'precision': 0.7426470588235294, 'recall': 0.5611111111111111, 'f1': 0.6392405063291139, 'number': 360}, 'overall_precision': 0.6125, 'overall_recall': 0.5726210350584308, 'overall_f1': 0.5918895599654875, 'overall_accuracy': 0.7669088457793424}
			------------EPOCH 20---------------
Loss:  tensor(32.0315, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.8808, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.3636, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.2215, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.8582, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.7512, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.7088, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.9920, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(133.2465, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(67.9095, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.5063, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.9340, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.6948, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(83.7739, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.5554, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.4255, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.1194, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.8542, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.3896, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.7612, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.4987, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.1877, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.1307, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.2078, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.2970, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.7317, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.6333, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.7517, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.8708, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.3495, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.8488, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.8385, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.7790, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.6595, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.8782, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.2369, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.4047, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.9159, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.2140, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.2652, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.7470, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.4710, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.5389, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.5478, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.1934, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.1501, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.8123, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(77.5583, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.5257, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(90.3505, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.5099, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(68.8026, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.5908, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.5398, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.7646, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.8813, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.3142, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.4021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.2446, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5913978494623656, 'recall': 0.4602510460251046, 'f1': 0.5176470588235295, 'number': 239}, 'P': {'precision': 0.6352357320099256, 'recall': 0.7111111111111111, 'f1': 0.6710353866317169, 'number': 360}, 'overall_precision': 0.6213921901528013, 'overall_recall': 0.6110183639398998, 'overall_f1': 0.6161616161616161, 'overall_accuracy': 0.7905025840760992}
			------------EPOCH 21---------------
Loss:  tensor(41.1027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.5609, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.9723, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.7919, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.6664, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.6918, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.5459, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.1539, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(153.5962, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(90.8413, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(71.1550, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.6153, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.9439, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(107.1090, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.8945, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.6209, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.9343, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.4629, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.3704, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.3835, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.8016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.4511, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.6711, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.0769, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.5554, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.5987, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.1681, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.1482, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.5317, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.7685, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.1560, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.9613, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.5325, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.3915, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.9483, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.0121, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.4701, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.1261, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.7685, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.4928, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.4744, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.6220, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.1934, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.7200, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(67.7288, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.6100, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.5274, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.0760, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.1782, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(75.5674, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.0337, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(64.8217, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.4259, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(101.7931, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(61.1435, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.4663, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.4154, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.2082, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.3319, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.49612403100775193, 'recall': 0.5355648535564853, 'f1': 0.5150905432595573, 'number': 239}, 'P': {'precision': 0.6198083067092651, 'recall': 0.5388888888888889, 'f1': 0.5765230312035661, 'number': 360}, 'overall_precision': 0.563922942206655, 'overall_recall': 0.5375626043405676, 'overall_f1': 0.5504273504273505, 'overall_accuracy': 0.7692307692307693}
			------------EPOCH 22---------------
Loss:  tensor(30.5575, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.9143, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.2995, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.6277, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.9862, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.2275, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.5741, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(61.7876, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(141.9222, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(72.9577, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.0592, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.7038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.2208, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(69.6158, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.5056, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.1231, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.3013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.1274, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.0680, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.5300, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.9655, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.9134, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.8075, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.3292, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.2244, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.7541, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.9690, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.4292, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.3019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.7857, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.7267, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.7988, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.7438, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.3160, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.3557, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.2643, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.3409, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.0447, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.5656, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.8845, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.4828, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.9213, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.2128, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.7269, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.3081, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.3565, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.3704, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.4380, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.9276, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.7979, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.5571, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.4589, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.9480, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.5131, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.3748, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.3620, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.7771, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.4327, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.5593, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4581818181818182, 'recall': 0.5271966527196653, 'f1': 0.49027237354085607, 'number': 239}, 'P': {'precision': 0.6203389830508474, 'recall': 0.5083333333333333, 'f1': 0.5587786259541985, 'number': 360}, 'overall_precision': 0.5421052631578948, 'overall_recall': 0.5158597662771286, 'overall_f1': 0.5286569717707442, 'overall_accuracy': 0.7523031982623024}
			------------EPOCH 23---------------
Loss:  tensor(33.9218, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.4956, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.4869, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.5077, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.3327, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.0743, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.5318, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.8559, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(133.2314, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(64.5248, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.8469, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.8613, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.7244, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(59.0441, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.5632, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.2134, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.5485, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.3532, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.2766, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.0690, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.1732, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.2234, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.6351, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.4671, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.7172, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.3122, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.5316, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.3602, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.8649, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.0095, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.7674, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.5625, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.8642, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.4602, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.4537, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.0342, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.8875, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.5956, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.1906, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.8937, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.0812, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.7229, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.3947, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.1490, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.1897, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.5630, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.5103, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.6109, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.5790, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.2012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.0386, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.2962, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.6223, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.2472, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.6654, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.5456, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.2216, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.4273, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.7544, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.49808429118773945, 'recall': 0.5439330543933054, 'f1': 0.5199999999999999, 'number': 239}, 'P': {'precision': 0.62751677852349, 'recall': 0.5194444444444445, 'f1': 0.5683890577507599, 'number': 360}, 'overall_precision': 0.5670840787119857, 'overall_recall': 0.5292153589315526, 'overall_f1': 0.547495682210708, 'overall_accuracy': 0.7569470451651562}
			------------EPOCH 24---------------
Loss:  tensor(29.6429, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.9166, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.2738, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.4390, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.8726, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.5245, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.6454, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.0870, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(119.6072, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.5428, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.6555, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.4540, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.4849, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.8127, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.6513, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.6103, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.9138, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.1881, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.3885, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.7421, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.5844, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.0771, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.0906, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.1929, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.2414, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.9653, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.4597, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.9496, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.0678, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.7879, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.0955, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.0486, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.4391, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.0272, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.6848, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.7351, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.5291, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.2794, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.7578, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.2425, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.8942, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.1405, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.1989, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.1862, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.1892, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.0363, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.0288, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.4079, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.9313, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.7536, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.7571, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.4833, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.5177, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.8326, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.6910, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.0718, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.6785, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.9946, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.8623, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.48872180451127817, 'recall': 0.5439330543933054, 'f1': 0.5148514851485148, 'number': 239}, 'P': {'precision': 0.6632996632996633, 'recall': 0.5472222222222223, 'f1': 0.5996955859969558, 'number': 360}, 'overall_precision': 0.5808170515097691, 'overall_recall': 0.5459098497495827, 'overall_f1': 0.5628227194492256, 'overall_accuracy': 0.7573215489476444}
			------------EPOCH 25---------------
Loss:  tensor(27.1770, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.0802, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.7412, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.5960, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.7005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.9621, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.6986, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.4147, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(111.8753, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.4581, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.5936, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.9070, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.5944, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.2223, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.5856, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.0527, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.1921, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.6179, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.1765, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.2431, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.0067, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.2472, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.1674, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.6375, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.9938, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.4547, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.4882, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.1829, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.3469, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.1722, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.5930, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.2523, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.1231, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.8494, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.5133, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.0388, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.6266, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.1785, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.0278, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.9720, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.3618, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.3307, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.2674, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.4207, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.6812, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.5591, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.9871, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.5902, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.9268, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.0809, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.8835, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.4470, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.8011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.8173, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.0217, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.5136, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.4841, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.0195, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.5831, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5038759689922481, 'recall': 0.5439330543933054, 'f1': 0.5231388329979879, 'number': 239}, 'P': {'precision': 0.6632996632996633, 'recall': 0.5472222222222223, 'f1': 0.5996955859969558, 'number': 360}, 'overall_precision': 0.5891891891891892, 'overall_recall': 0.5459098497495827, 'overall_f1': 0.5667244367417679, 'overall_accuracy': 0.7591940678600854}
			------------EPOCH 26---------------
Loss:  tensor(25.3145, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.9861, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.1790, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.8896, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.5195, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.0562, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.4638, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.3420, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(105.1801, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.2799, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.8084, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.0059, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.2369, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.0440, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.3510, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.6938, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.4151, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.6992, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.5708, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.0386, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.6929, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.4789, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.5181, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.4214, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.2495, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.6406, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.1764, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.6362, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.6928, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.7730, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.8843, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.8988, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.3623, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.9746, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.6281, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.9126, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.9762, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.4302, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.6164, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.2080, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.4326, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.7359, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.8845, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.4082, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.1599, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.0906, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.2189, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.7903, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.3910, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.1878, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.5359, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.2376, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.4291, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.4157, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.2175, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.4284, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.9418, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.3169, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.8206, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4980694980694981, 'recall': 0.5397489539748954, 'f1': 0.5180722891566265, 'number': 239}, 'P': {'precision': 0.6632653061224489, 'recall': 0.5416666666666666, 'f1': 0.5963302752293578, 'number': 360}, 'overall_precision': 0.5858951175406871, 'overall_recall': 0.5409015025041736, 'overall_f1': 0.5625, 'overall_accuracy': 0.7560482360871845}
			------------EPOCH 27---------------
Loss:  tensor(23.8427, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.7251, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.9005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.6864, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.6779, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.1478, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.5940, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.8378, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(99.0263, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.4341, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.2422, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.6736, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.2591, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.4768, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.7228, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.4797, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.0821, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.2088, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.5285, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.1288, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.6399, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.7179, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.0577, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.5955, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.6723, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.7309, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.2247, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.1760, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.2244, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.6605, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.6065, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.8356, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.1196, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.4013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.9029, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.0723, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.4852, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.7539, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.6369, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.8472, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.9003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.5933, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.0707, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.3424, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.7392, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.7011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.5769, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.3799, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.0109, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.2445, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.0316, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.6740, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.4720, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.5360, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.4953, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.8308, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.7101, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.6897, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.4760, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5096525096525096, 'recall': 0.5523012552301255, 'f1': 0.5301204819277109, 'number': 239}, 'P': {'precision': 0.6440677966101694, 'recall': 0.5277777777777778, 'f1': 0.5801526717557252, 'number': 360}, 'overall_precision': 0.5812274368231047, 'overall_recall': 0.5375626043405676, 'overall_f1': 0.5585429314830876, 'overall_accuracy': 0.7526777020447907}
			------------EPOCH 28---------------
Loss:  tensor(20.9779, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.1169, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.7825, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.9071, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.0635, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.3528, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.0992, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.5699, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(90.5522, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.9491, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.1021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.7388, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.7455, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.4730, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.1386, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.0795, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.8972, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.5035, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.7509, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.4775, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.3855, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.9007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.7330, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.1594, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.4670, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.7416, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.8226, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.7796, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.7329, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.6308, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.1058, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.2125, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.9503, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.2028, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.4331, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.5414, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.0669, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.1270, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.7418, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.9357, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.9936, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.8605, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.3540, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.7340, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.0633, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.3514, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.1847, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.4952, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.5338, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.3483, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.9419, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.5775, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.7437, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.9784, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.3168, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.6666, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.9625, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.1024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.1952, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5115384615384615, 'recall': 0.5564853556485355, 'f1': 0.533066132264529, 'number': 239}, 'P': {'precision': 0.6474576271186441, 'recall': 0.5305555555555556, 'f1': 0.5832061068702289, 'number': 360}, 'overall_precision': 0.5837837837837838, 'overall_recall': 0.5409015025041736, 'overall_f1': 0.561525129982669, 'overall_accuracy': 0.7541008164182458}
			------------EPOCH 29---------------
Loss:  tensor(17.3260, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.7010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.7147, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.0097, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.2734, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.2648, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.6620, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.7347, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(88.3543, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.4556, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.3943, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.9104, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.2399, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.1743, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.9626, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.3540, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.3214, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.1392, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.2641, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.2537, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.8763, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.1627, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.6776, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.6569, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.1090, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.2495, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.4897, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.2891, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.4567, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.5869, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.7847, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.0387, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.4248, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.3829, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.0383, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.7638, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.1777, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.1150, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.8246, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.2743, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.3098, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.6107, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.1179, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.7133, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.1753, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.1274, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.3760, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.9388, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.3457, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.6070, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.7359, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.6904, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.8066, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.0272, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.7661, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.3358, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.3564, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.5963, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.6250, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5057471264367817, 'recall': 0.5523012552301255, 'f1': 0.528, 'number': 239}, 'P': {'precision': 0.6577181208053692, 'recall': 0.5444444444444444, 'f1': 0.5957446808510638, 'number': 360}, 'overall_precision': 0.5867620751341681, 'overall_recall': 0.5475792988313857, 'overall_f1': 0.5664939550949915, 'overall_accuracy': 0.7570968466781515}
			------------EPOCH 30---------------
Loss:  tensor(13.6048, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.7503, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.3629, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.9336, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.8108, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.4356, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.8111, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.5090, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(79.5033, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.7838, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.5679, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.4367, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.0230, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.4723, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.4683, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.0579, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.3308, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.4635, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.7906, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.1530, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.7299, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.0500, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.8506, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.2289, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.0692, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.3659, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.3954, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.7143, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.1174, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.2515, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.4643, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.3471, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.0719, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.0576, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.2942, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.5996, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.9218, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.4331, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.7445, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.1532, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.7301, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.2110, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.6647, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.5391, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.3358, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.7274, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.9424, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.4927, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.2969, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.8285, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.1359, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.8223, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.1848, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.4329, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.5575, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.7481, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.6828, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.3590, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.4865, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5057471264367817, 'recall': 0.5523012552301255, 'f1': 0.528, 'number': 239}, 'P': {'precision': 0.6612377850162866, 'recall': 0.5638888888888889, 'f1': 0.608695652173913, 'number': 360}, 'overall_precision': 0.5897887323943662, 'overall_recall': 0.5592654424040067, 'overall_f1': 0.5741216795201372, 'overall_accuracy': 0.7660849374578683}


		-------------RUN 3-----------
			------------EPOCH 1---------------
Loss:  tensor(2352.8599, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2811.7480, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1702.7783, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1685.7734, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2438.6128, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1817.5220, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3484.7798, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1695.1267, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2313.8403, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2449.7061, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3999.7385, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2978.7244, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2185.7117, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1540.1715, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(510.0768, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2692.1997, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1272.7994, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2458.0449, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2760.0847, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2279.5312, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2258.1108, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1317.0901, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1696.7910, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2067.2212, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1748.0243, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2195.3774, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1513.1702, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2549.5054, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2562.2258, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2719.2375, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1027.6875, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1897.5510, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3163.9563, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1681.1876, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1210.5219, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2288.6885, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1162.0011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1668.8882, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1074.1920, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1732.0681, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2124.3274, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1241.0934, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1759.4490, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1138.6588, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1873.0791, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1043.2949, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2165.7671, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2486.4756, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(830.0471, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1270.2222, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2289.0508, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2399.1111, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2403.6680, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2426.3716, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2119.4500, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(906.7227, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1079.2827, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.09558823529411764, 'recall': 0.10970464135021098, 'f1': 0.10216110019646366, 'number': 237}, 'P': {'precision': 0.18823529411764706, 'recall': 0.12598425196850394, 'f1': 0.1509433962264151, 'number': 381}, 'overall_precision': 0.14041745730550284, 'overall_recall': 0.11974110032362459, 'overall_f1': 0.12925764192139738, 'overall_accuracy': 0.5564313616071429}
			------------EPOCH 2---------------
Loss:  tensor(1327.2930, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1594.4508, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1139.9315, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1118.5267, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1666.0205, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1213.6937, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2373.3242, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1122.3540, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1589.1816, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1498.6903, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2747.9861, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2014.2166, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1459.9996, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1146.0221, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(387.9320, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1742.0261, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(841.4959, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1709.6199, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1900.7897, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1457.0199, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2037.1594, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1214.7338, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1646.7830, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2060.0544, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1530.9987, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1860.6152, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1058.6772, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1737.8378, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1968.3171, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2217.2319, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(736.4135, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1492.2562, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2609.2959, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1191.1118, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(912.4406, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1765.9968, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(970.7930, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1405.3433, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(889.5762, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1220.8438, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1663.3389, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1033.9741, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1329.0840, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(846.4431, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1472.8234, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(818.1726, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1739.1790, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1766.2928, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(620.5017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1084.3005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1898.3693, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1922.5154, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2034.6494, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2082.4910, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1797.5527, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(726.8962, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(891.6013, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.07612456747404844, 'recall': 0.09282700421940929, 'f1': 0.08365019011406845, 'number': 237}, 'P': {'precision': 0.3469945355191257, 'recall': 0.3333333333333333, 'f1': 0.34002677376171353, 'number': 381}, 'overall_precision': 0.22748091603053436, 'overall_recall': 0.24110032362459546, 'overall_f1': 0.2340926944226237, 'overall_accuracy': 0.666015625}
			------------EPOCH 3---------------
Loss:  tensor(993.3188, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1228.2423, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(929.9176, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(914.4014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1348.7502, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(998.4648, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1985.9664, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(961.3831, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1352.6903, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1236.3571, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2338.6165, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1710.0640, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1226.5958, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(994.0174, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(329.7881, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1438.9318, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(695.4670, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1398.6296, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1529.5991, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1079.9158, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1487.0366, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(909.7705, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1129.0461, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1484.6859, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1400.6223, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1702.6140, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(886.4034, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1505.7970, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1497.3917, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1799.5344, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(559.5783, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1104.9370, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1978.9764, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(860.3212, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(694.6044, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1427.4082, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(861.0419, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1228.0424, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(747.3636, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1249.8013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1426.6885, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(886.9679, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1257.5980, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(739.9768, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1249.2811, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(720.3702, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1384.4237, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1356.1410, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(487.2396, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(927.0326, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1515.1592, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1526.6721, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1821.5394, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1846.1313, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1567.3345, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(606.9297, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(708.1052, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.08801955990220049, 'recall': 0.1518987341772152, 'f1': 0.11145510835913314, 'number': 237}, 'P': {'precision': 0.3731060606060606, 'recall': 0.5170603674540682, 'f1': 0.43344334433443343, 'number': 381}, 'overall_precision': 0.2486659551760939, 'overall_recall': 0.37702265372168287, 'overall_f1': 0.2996784565916399, 'overall_accuracy': 0.6654575892857143}
			------------EPOCH 4---------------
Loss:  tensor(763.0225, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(910.1052, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(767.9717, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(764.5388, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1103.5298, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(794.9025, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1769.5392, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(718.5664, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1094.8789, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1076.1179, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1861.9957, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1507.4204, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1009.2905, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(895.6879, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(285.6821, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1166.5212, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(551.5672, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1158.2716, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1259.4358, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(780.7047, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1046.5198, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(570.3334, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(652.6547, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(962.6053, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1034.1594, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1346.7324, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(785.7687, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1199.9526, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1169.5651, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1505.0552, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(453.2061, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(838.9788, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1566.4073, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(676.1230, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(522.5352, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1195.1277, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(774.4762, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(876.9474, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(571.8734, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(809.9972, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1135.9543, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(704.8717, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(969.7322, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(564.2881, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1136.2080, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(614.1517, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1134.7131, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1155.5675, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(386.8586, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(862.7625, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1375.9050, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1201.7920, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1396.7344, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1499.7227, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1259.7228, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(429.2951, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(488.4801, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.18597560975609756, 'recall': 0.25738396624472576, 'f1': 0.21592920353982303, 'number': 237}, 'P': {'precision': 0.4883268482490272, 'recall': 0.6587926509186351, 'f1': 0.5608938547486033, 'number': 381}, 'overall_precision': 0.37054631828978624, 'overall_recall': 0.5048543689320388, 'overall_f1': 0.4273972602739726, 'overall_accuracy': 0.6764787946428571}
			------------EPOCH 5---------------
Loss:  tensor(563.9552, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(698.0870, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(659.8254, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(679.0165, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1024.9618, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(705.1702, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1784.0344, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(641.6741, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(994.9684, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(977.2236, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1651.5613, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1390.8906, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(724.9263, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(818.9714, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(277.0607, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1034.1445, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(459.9508, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(926.6375, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1122.4004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(540.9079, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(729.6571, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(400.0352, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(432.3168, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(745.4030, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(875.0486, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1248.0192, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(708.0782, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1070.3965, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(949.6284, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1242.4926, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(344.2228, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(596.1091, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1230.6594, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(497.7755, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(350.7286, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(930.5399, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(631.7957, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(677.7786, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(405.2482, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(600.4482, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(930.2523, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(553.9205, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(803.3225, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(435.1251, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(872.1414, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(399.4198, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(799.5254, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(804.7174, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(276.5474, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(581.1337, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(791.8745, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(797.2764, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(954.0835, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1089.0930, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1025.2014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(336.6474, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(372.5299, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3888888888888889, 'recall': 0.5316455696202531, 'f1': 0.44919786096256686, 'number': 237}, 'P': {'precision': 0.5787781350482315, 'recall': 0.47244094488188976, 'f1': 0.5202312138728323, 'number': 381}, 'overall_precision': 0.4818897637795276, 'overall_recall': 0.49514563106796117, 'overall_f1': 0.48842777334397447, 'overall_accuracy': 0.7141462053571429}
			------------EPOCH 6---------------
Loss:  tensor(359.5743, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(434.4925, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(514.9966, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(516.5992, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(662.5888, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(471.3357, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1145.5890, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(495.8716, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(605.4152, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(723.9757, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1165.6125, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(814.5015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(487.7240, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(557.3293, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(177.6702, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1054.3115, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(306.1363, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(623.8388, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(951.1000, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(399.9681, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(782.5909, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(391.5677, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(434.4507, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(747.2726, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(963.0137, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1514.1093, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(552.9116, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1040.7766, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(687.1419, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(994.4692, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(277.1530, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(451.5958, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1001.4974, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(355.8356, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(272.7977, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(662.9059, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(592.1373, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(874.0769, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(403.5717, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(759.5067, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(809.5659, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(626.9868, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1205.4863, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(665.9453, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(922.4962, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(400.3923, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(772.8688, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(773.3739, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(212.9525, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(444.9443, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(583.2728, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(655.9954, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(688.5886, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(754.2700, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(724.1794, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(218.2733, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(299.1357, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.2793103448275862, 'recall': 0.6835443037974683, 'f1': 0.39657282741738065, 'number': 237}, 'P': {'precision': 0.5158730158730159, 'recall': 0.17060367454068243, 'f1': 0.25641025641025644, 'number': 381}, 'overall_precision': 0.32152974504249293, 'overall_recall': 0.3673139158576052, 'overall_f1': 0.34290030211480366, 'overall_accuracy': 0.5770786830357143}
			------------EPOCH 7---------------
Loss:  tensor(353.4501, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(408.8655, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(648.9726, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(668.7657, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(688.7975, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(616.9714, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1369.7623, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(756.2972, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(608.2776, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(583.0134, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(937.8926, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(785.3821, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(387.9921, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(379.7590, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(140.4658, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(787.8016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(320.4213, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(623.8844, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(880.5459, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(299.7915, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(494.2257, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(585.1295, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(493.0765, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(892.4493, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1302.1953, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1390.1394, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(740.2571, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1366.4478, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(993.3461, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1120.7112, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(302.7883, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(505.9788, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(787.5144, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(316.1590, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(208.0614, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(659.6852, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(444.8357, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(523.8395, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(233.4531, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(473.9922, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(649.0444, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(468.3659, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(652.8545, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(285.2421, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(955.3073, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(401.2919, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(998.1621, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1313.8987, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(218.5133, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(517.6549, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(855.3552, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(924.7454, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(575.0088, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(627.2537, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(664.6477, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(242.1944, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(218.5890, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4205128205128205, 'recall': 0.3459915611814346, 'f1': 0.3796296296296296, 'number': 237}, 'P': {'precision': 0.5185185185185185, 'recall': 0.5511811023622047, 'f1': 0.5343511450381679, 'number': 381}, 'overall_precision': 0.4866666666666667, 'overall_recall': 0.47249190938511326, 'overall_f1': 0.4794745484400657, 'overall_accuracy': 0.7253766741071429}
			------------EPOCH 8---------------
Loss:  tensor(295.5980, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(396.1883, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(522.3052, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(372.2632, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(687.2609, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(470.8100, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1037.2466, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(524.5183, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(553.1902, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(466.0375, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1049.4827, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(654.7585, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(380.0219, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(394.6474, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(106.0610, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(732.3989, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(253.6555, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(310.3846, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(657.8110, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(238.4678, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(316.7469, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(211.0550, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(261.6524, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(358.1707, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(537.0675, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(718.6324, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(368.5804, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(843.1627, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(493.8583, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(764.3662, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(236.2061, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(496.1644, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(884.2832, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(286.6906, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(191.4334, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(598.8840, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(331.8976, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(406.7683, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(165.0927, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(405.9541, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(489.3469, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(276.9790, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(493.0866, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(182.1128, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(536.1458, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(253.7263, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(322.2329, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(279.1085, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(122.1555, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(248.2741, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(521.1406, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(471.7435, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(560.8143, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(519.0698, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(575.8640, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(190.6214, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(120.4284, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4426877470355731, 'recall': 0.47257383966244726, 'f1': 0.45714285714285713, 'number': 237}, 'P': {'precision': 0.6847826086956522, 'recall': 0.49606299212598426, 'f1': 0.5753424657534246, 'number': 381}, 'overall_precision': 0.5689981096408318, 'overall_recall': 0.48705501618122976, 'overall_f1': 0.5248474280732345, 'overall_accuracy': 0.7017299107142857}
			------------EPOCH 9---------------
Loss:  tensor(133.7813, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(198.4059, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(364.8290, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(286.2869, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(375.8607, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(240.2147, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(547.7899, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(243.4339, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(391.5190, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(401.7109, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(468.2969, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(355.0193, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(190.0382, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(313.4922, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(155.9487, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(506.8077, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(248.0178, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(192.2833, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(549.2582, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(170.8780, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(207.6347, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(139.8606, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(173.5388, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(189.8043, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(511.9678, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(501.9189, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(246.7315, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(656.9388, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(316.4434, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(556.2934, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(221.0268, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(336.4110, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(583.3756, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(182.3371, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(203.1540, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(566.2757, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(540.2609, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(518.1969, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(262.9220, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(517.6423, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(671.9529, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(240.6129, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(408.5877, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(216.2172, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(265.9024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(171.6648, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(208.1570, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(131.0622, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(72.2334, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(126.4412, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(248.3031, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(363.9348, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(417.5256, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(247.2291, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(353.3558, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(133.7822, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(158.1842, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.336734693877551, 'recall': 0.27848101265822783, 'f1': 0.30484988452655887, 'number': 237}, 'P': {'precision': 0.6081081081081081, 'recall': 0.23622047244094488, 'f1': 0.34026465028355385, 'number': 381}, 'overall_precision': 0.45348837209302323, 'overall_recall': 0.2524271844660194, 'overall_f1': 0.3243243243243243, 'overall_accuracy': 0.5583147321428571}
			------------EPOCH 10---------------
Loss:  tensor(362.1717, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(797.6992, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(921.6503, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(589.2289, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(590.5367, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(405.4913, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(966.3093, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(284.5316, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(295.9926, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(223.0104, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(605.8387, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(555.1246, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(143.4506, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(184.8694, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.2276, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(300.9020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(163.0477, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(99.8229, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(390.1323, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(87.3401, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(161.6140, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(140.4564, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(141.0617, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(219.6741, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(522.4808, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(564.4191, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(233.3926, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(661.1331, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(277.2363, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(685.2342, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(141.0975, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(550.0219, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(968.4740, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(236.2311, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(290.1969, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(726.9723, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(840.4275, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1042.6221, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(533.3633, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(741.9133, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1184.7333, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(715.2742, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(983.8056, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(222.2083, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(502.4256, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(188.6304, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(685.6724, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(223.9097, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(116.1000, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(266.9695, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(451.8095, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(478.8497, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(277.1303, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(277.4906, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(348.4046, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(107.3428, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(116.6179, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.38549618320610685, 'recall': 0.42616033755274263, 'f1': 0.4048096192384769, 'number': 237}, 'P': {'precision': 0.5247148288973384, 'recall': 0.36220472440944884, 'f1': 0.42857142857142855, 'number': 381}, 'overall_precision': 0.4552380952380952, 'overall_recall': 0.38673139158576053, 'overall_f1': 0.41819772528433946, 'overall_accuracy': 0.6692940848214286}
			------------EPOCH 11---------------
Loss:  tensor(92.0345, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(198.4320, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(269.2869, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(174.8562, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(452.3335, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(366.5239, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(784.9360, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(529.4686, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(772.2050, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(585.0559, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1509.4099, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1298.5497, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(384.7802, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(387.2007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(96.8728, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(922.9679, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(278.5731, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(290.6911, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(441.9188, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(431.2247, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(193.6990, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(102.3703, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(141.6938, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(151.8817, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(318.0995, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(331.5425, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(238.3876, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(499.9652, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(198.1838, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(301.1810, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(75.2506, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(192.8358, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(328.8231, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(94.0011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(87.9402, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(277.4309, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(256.0650, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(290.4222, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(110.8704, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(223.3572, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(491.1617, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(363.9506, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(490.0162, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(156.9641, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(456.4185, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(152.4300, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(473.9913, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(249.7597, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(102.3430, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(282.1426, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(678.8864, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(531.6407, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(412.4072, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(381.5478, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(569.1772, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(202.8535, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(98.6067, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.44744744744744747, 'recall': 0.6286919831223629, 'f1': 0.5228070175438596, 'number': 237}, 'P': {'precision': 0.6955380577427821, 'recall': 0.6955380577427821, 'f1': 0.6955380577427821, 'number': 381}, 'overall_precision': 0.5798319327731093, 'overall_recall': 0.6699029126213593, 'overall_f1': 0.6216216216216217, 'overall_accuracy': 0.7430245535714286}
			------------EPOCH 12---------------
Loss:  tensor(77.9271, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(113.4776, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(266.2957, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(137.0290, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(321.3242, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(238.7093, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(423.3723, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(172.7054, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(268.1270, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(215.7051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(301.7743, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(429.9861, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(153.4263, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(174.6701, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(68.3272, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(189.1134, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(157.8139, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(83.3981, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(214.1281, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(113.6580, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(108.6110, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(101.1131, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(92.3794, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(123.6223, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(245.9796, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(325.1774, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(307.8016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(459.9346, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(317.4464, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(425.6645, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(86.5300, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(259.1427, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(363.9550, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(113.0662, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(133.1698, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(264.9624, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(136.6427, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(223.8418, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.1860, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(217.8242, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(201.1439, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(157.1669, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(276.9998, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(62.4528, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(165.3273, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(81.0207, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(89.2977, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(71.0399, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.3469, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(83.2302, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(107.5114, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(110.7201, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(147.1552, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(123.8497, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(149.3122, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(83.5963, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(59.6116, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5023255813953489, 'recall': 0.45569620253164556, 'f1': 0.4778761061946903, 'number': 237}, 'P': {'precision': 0.6104166666666667, 'recall': 0.7690288713910761, 'f1': 0.6806039488966319, 'number': 381}, 'overall_precision': 0.576978417266187, 'overall_recall': 0.6488673139158576, 'overall_f1': 0.6108149276466107, 'overall_accuracy': 0.7658342633928571}
			------------EPOCH 13---------------
Loss:  tensor(55.4167, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(72.1033, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(157.6588, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(108.2007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(241.1317, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(145.7700, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(313.4714, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(76.7233, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(166.4791, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(122.5827, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(210.5411, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(246.5532, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(91.5642, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(110.7291, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.9643, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(142.1398, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(96.2033, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(61.8065, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(211.3412, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(74.3422, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(110.4747, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(62.6749, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(68.1119, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(91.9124, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(167.3904, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(274.7020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(157.1100, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(289.3961, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(109.1982, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(232.0122, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(87.1739, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(168.1250, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(221.5647, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.6278, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.0269, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(193.3036, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(68.7288, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(126.5951, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.3555, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(116.6600, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(141.5971, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(84.1233, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(139.7477, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.0579, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(89.9942, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.8057, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(69.0379, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.9423, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.1227, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(59.2866, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.9333, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(71.3093, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(83.7990, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(90.5215, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(86.5827, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.0738, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.4936, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5253164556962026, 'recall': 0.350210970464135, 'f1': 0.420253164556962, 'number': 237}, 'P': {'precision': 0.5873015873015873, 'recall': 0.7769028871391076, 'f1': 0.6689265536723165, 'number': 381}, 'overall_precision': 0.5725075528700906, 'overall_recall': 0.6132686084142395, 'overall_f1': 0.5921875, 'overall_accuracy': 0.7516043526785714}
			------------EPOCH 14---------------
Loss:  tensor(59.3168, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.9048, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(128.3510, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.9428, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(166.5449, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(113.9925, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(167.9535, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(70.1481, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(241.2000, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(102.8073, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(295.4324, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(189.3459, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(150.5977, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(163.4723, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.5429, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(221.8024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(153.9473, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.4835, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(110.3985, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.0689, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(59.6667, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.7477, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.5800, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(64.1564, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(163.8405, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(146.5307, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(142.5226, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(144.4621, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.8361, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(115.6860, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.1466, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(128.7109, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(259.9648, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.6886, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.7723, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(189.0758, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(131.6695, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(279.7287, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(124.2259, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(231.9124, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(152.9523, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(111.1110, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(177.6892, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.3391, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(130.7255, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(75.4055, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(86.1115, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.1103, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.7937, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.0029, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.9553, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(93.0200, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(70.0678, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(78.9008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(67.2985, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.0171, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.3730, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5384615384615384, 'recall': 0.41350210970464135, 'f1': 0.46778042959427213, 'number': 237}, 'P': {'precision': 0.6092436974789915, 'recall': 0.7611548556430446, 'f1': 0.676779463243874, 'number': 381}, 'overall_precision': 0.5896656534954408, 'overall_recall': 0.627831715210356, 'overall_f1': 0.6081504702194357, 'overall_accuracy': 0.7568359375}
			------------EPOCH 15---------------
Loss:  tensor(44.0621, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.3910, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(66.9053, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.6838, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(119.0648, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(77.7275, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(148.3074, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.0074, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(181.5518, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(89.3491, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(169.0619, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(166.7033, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(76.5891, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(121.0721, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.3774, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(143.6984, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(173.5867, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(252.1517, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(192.4176, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.8654, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(71.6678, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(182.8438, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(71.1371, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(245.9988, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(269.3865, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(141.4628, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(181.2918, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(287.7687, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.0350, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(155.6683, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.3692, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(155.8256, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(100.1028, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.1134, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.2468, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(120.4504, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.4343, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(81.8766, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.8984, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(61.2374, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(98.7335, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(88.4248, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(97.3876, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.1478, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(208.4427, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(173.8388, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(336.7551, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(228.2782, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(100.4864, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(418.8633, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(490.2117, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(757.3893, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(220.0703, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(144.3558, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(214.6711, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(66.2871, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.1105, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4981549815498155, 'recall': 0.569620253164557, 'f1': 0.531496062992126, 'number': 237}, 'P': {'precision': 0.6532663316582915, 'recall': 0.6824146981627297, 'f1': 0.6675224646983312, 'number': 381}, 'overall_precision': 0.5904334828101644, 'overall_recall': 0.63915857605178, 'overall_f1': 0.6138306138306138, 'overall_accuracy': 0.7606026785714286}
			------------EPOCH 16---------------
Loss:  tensor(56.6718, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.3779, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(61.1568, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.0035, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(114.2243, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(66.4310, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(105.8445, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.2657, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(105.1971, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.2314, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(141.8365, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(65.0329, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(85.2700, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(103.8187, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.5369, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(93.3069, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(107.6667, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(73.3702, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(178.7642, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(121.6057, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(167.5210, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(95.3247, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(91.3321, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(263.2746, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(596.0569, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(497.4694, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(345.6093, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(750.3062, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1030.5947, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(800.8341, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(101.0492, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(278.9070, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(485.9541, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(204.9050, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(82.0912, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(519.4106, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(84.0599, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(83.0587, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.6024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(83.6462, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(82.3537, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(85.5283, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(94.6176, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.7005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(83.9636, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.2520, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(85.3719, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(90.3582, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.6886, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(62.5664, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(91.0134, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(193.6890, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(434.5335, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(557.9996, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(712.6814, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(138.0946, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(249.0056, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.30514096185737977, 'recall': 0.7763713080168776, 'f1': 0.43809523809523804, 'number': 237}, 'P': {'precision': 0.7589285714285714, 'recall': 0.2230971128608924, 'f1': 0.3448275862068966, 'number': 381}, 'overall_precision': 0.37622377622377623, 'overall_recall': 0.43527508090614886, 'overall_f1': 0.4036009002250563, 'overall_accuracy': 0.6047712053571429}
			------------EPOCH 17---------------
Loss:  tensor(291.6227, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(316.9465, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(293.5839, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(418.4317, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(389.5430, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(285.8839, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(926.0925, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(248.8203, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(369.1776, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(350.0448, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(326.7984, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(302.5902, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(88.7640, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(169.5135, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.0126, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(139.2410, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(109.4905, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.9796, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(115.0991, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(72.1537, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(80.7728, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(59.6971, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.9627, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(79.8485, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(133.1358, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(118.0766, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(99.6940, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(173.7371, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(146.6082, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(189.7996, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.6401, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(124.9700, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(316.9121, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(235.1198, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(129.9085, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(318.9976, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(251.7539, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(327.3588, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(244.9949, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(358.5921, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(344.1720, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(372.3326, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(611.0165, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(411.5059, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(325.7673, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(105.6787, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(177.8666, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(164.6607, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.5094, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(128.4846, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(107.9233, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(211.0323, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(125.3417, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(187.0871, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(135.6102, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(69.9409, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.9668, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.44808743169398907, 'recall': 0.6919831223628692, 'f1': 0.5439469320066335, 'number': 237}, 'P': {'precision': 0.7333333333333333, 'recall': 0.4330708661417323, 'f1': 0.5445544554455446, 'number': 381}, 'overall_precision': 0.55668358714044, 'overall_recall': 0.5323624595469255, 'overall_f1': 0.5442514474772538, 'overall_accuracy': 0.7064732142857143}
			------------EPOCH 18---------------
Loss:  tensor(51.3343, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.5537, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(73.1093, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.3408, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(167.5709, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.1413, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(229.3455, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(82.3324, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(200.9288, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(117.5449, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(185.5799, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(223.3293, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(150.5823, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(289.7735, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.2294, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(398.9447, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(288.2407, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(149.5844, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(604.4233, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(325.6186, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(84.7367, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(77.2966, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(126.0688, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(108.4918, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(147.3339, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(146.9557, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(142.7235, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(158.9568, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(88.1495, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(154.6552, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.3546, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(168.9509, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(128.3821, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.5701, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.5297, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(124.6449, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.4443, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(77.4672, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.5472, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(65.3078, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(119.2944, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(68.6825, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(131.7318, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.5254, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(92.9394, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.2983, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(104.5262, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(67.2421, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.9153, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.2855, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(122.5600, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(155.8889, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(261.9759, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(310.0942, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(224.4711, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.1245, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.9399, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.43043478260869567, 'recall': 0.4177215189873418, 'f1': 0.423982869379015, 'number': 237}, 'P': {'precision': 0.6086956521739131, 'recall': 0.8083989501312336, 'f1': 0.6944757609921083, 'number': 381}, 'overall_precision': 0.5529891304347826, 'overall_recall': 0.6585760517799353, 'overall_f1': 0.601181683899557, 'overall_accuracy': 0.7599748883928571}
			------------EPOCH 19---------------
Loss:  tensor(75.5893, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.4575, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(100.5334, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(66.5458, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(148.2515, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(97.8050, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(216.3520, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(81.1409, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(108.0678, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(65.1759, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(147.7424, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(73.1137, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.5097, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(81.9953, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.9063, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(70.6614, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(67.3872, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(90.9979, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(126.9484, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(71.1501, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(74.1113, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.7663, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.1095, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(68.2898, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(90.1400, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(104.0757, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(74.1085, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(110.7905, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.8454, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(88.2655, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.5502, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(112.7318, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(88.4980, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.4242, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.2633, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(108.5718, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.5904, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(69.3082, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.6700, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.1597, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.3357, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.1364, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.9006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.5733, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.8454, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.8478, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.3089, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.9106, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.3688, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.1250, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.4476, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.9824, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(68.4731, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(78.4347, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.3382, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.7328, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.3311, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4588235294117647, 'recall': 0.6582278481012658, 'f1': 0.5407279029462738, 'number': 237}, 'P': {'precision': 0.7238095238095238, 'recall': 0.5984251968503937, 'f1': 0.6551724137931035, 'number': 381}, 'overall_precision': 0.5862595419847328, 'overall_recall': 0.6213592233009708, 'overall_f1': 0.6032992930086409, 'overall_accuracy': 0.7571149553571429}
			------------EPOCH 20---------------
Loss:  tensor(34.1608, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.2236, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.2314, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.2508, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(84.5200, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.8752, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(114.8727, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.7406, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(82.1946, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.4974, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(88.7511, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.9568, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.3253, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(75.3921, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.5364, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(61.5647, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(71.9752, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.9041, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(71.1257, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.6840, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.4043, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.6609, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.1902, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.6196, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(82.2595, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(83.6860, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(76.2655, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(112.7478, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.8511, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(82.0316, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.6635, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(83.5798, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(80.7454, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.7848, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.6592, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(101.0358, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.8527, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.3589, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.1524, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.1000, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.6014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.0238, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.6056, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.5105, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.2610, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.3881, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.5937, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.6747, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.0673, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.1084, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.0705, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.2629, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.8913, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.7435, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.5854, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.0191, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.7850, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4230769230769231, 'recall': 0.6033755274261603, 'f1': 0.4973913043478261, 'number': 237}, 'P': {'precision': 0.6716417910447762, 'recall': 0.5905511811023622, 'f1': 0.6284916201117319, 'number': 381}, 'overall_precision': 0.5468053491827637, 'overall_recall': 0.5954692556634305, 'overall_f1': 0.5701006971340047, 'overall_accuracy': 0.7596958705357143}
			------------EPOCH 21---------------
Loss:  tensor(30.7226, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.5979, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.3701, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.3504, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(64.6942, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.1980, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(75.7749, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.5421, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(65.2079, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.6670, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(90.4503, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.7588, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.5039, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(78.7309, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.7244, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.3824, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(75.0140, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.0586, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.4318, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.5718, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.4076, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.0489, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.5424, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.3620, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(75.0409, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(64.0301, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.3866, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(83.4283, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.2983, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.8189, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.7753, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(120.5020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.9357, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.8420, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.5930, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(112.6633, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.5850, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.0432, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.8271, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.2711, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.3004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.3766, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.7930, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.1431, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.2107, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.3607, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.5106, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.0553, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.4376, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.4671, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.0340, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.6487, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.9532, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.9549, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.0881, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.7149, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.7185, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.46689895470383275, 'recall': 0.5654008438818565, 'f1': 0.5114503816793893, 'number': 237}, 'P': {'precision': 0.6641414141414141, 'recall': 0.6902887139107612, 'f1': 0.676962676962677, 'number': 381}, 'overall_precision': 0.5812591508052709, 'overall_recall': 0.6423948220064725, 'overall_f1': 0.6102997694081477, 'overall_accuracy': 0.7799246651785714}
			------------EPOCH 22---------------
Loss:  tensor(31.6135, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.8349, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.4801, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.7274, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.7958, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.9723, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(59.0141, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.5295, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.0304, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.9267, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(68.8055, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.4194, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.0362, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(59.3080, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.2732, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.3920, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.4648, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.8128, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.2676, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.5430, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.0129, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.8642, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.7915, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.7708, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.5046, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.7513, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.2355, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.8537, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.0809, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.2777, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.2163, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(71.4718, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.3984, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.1598, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.1262, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(72.9268, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.6894, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.9673, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.7919, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.0294, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.2142, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.8450, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.4035, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.9754, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.2786, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.4449, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.8731, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.0463, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.1818, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.9104, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.6640, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.0098, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.6087, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.8607, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.0334, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.0457, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.8349, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4641638225255973, 'recall': 0.5738396624472574, 'f1': 0.5132075471698113, 'number': 237}, 'P': {'precision': 0.6675257731958762, 'recall': 0.6797900262467191, 'f1': 0.6736020806241872, 'number': 381}, 'overall_precision': 0.580029368575624, 'overall_recall': 0.63915857605178, 'overall_f1': 0.6081601231716705, 'overall_accuracy': 0.7755998883928571}
			------------EPOCH 23---------------
Loss:  tensor(23.9222, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.4272, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.8763, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.7935, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.3794, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.5484, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.1165, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.7749, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.8159, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.1700, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.7110, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.6805, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.9982, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.3776, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.2223, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.1091, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.3426, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.4204, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.8518, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.4150, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.5738, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.3030, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.3174, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.4325, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.0801, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.8149, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.4905, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.3613, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.0963, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.7015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.0856, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(70.8710, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.2272, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.3869, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.8831, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.8109, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.2624, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.3226, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.3783, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.7888, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.5533, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.5375, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.3592, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.4246, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.2730, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.5540, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.1237, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.5544, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.9170, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.3881, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.1492, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.6264, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.5474, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.3802, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.3653, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.5466, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.1818, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.46619217081850534, 'recall': 0.5527426160337553, 'f1': 0.5057915057915058, 'number': 237}, 'P': {'precision': 0.6770025839793282, 'recall': 0.6876640419947506, 'f1': 0.6822916666666666, 'number': 381}, 'overall_precision': 0.5883233532934131, 'overall_recall': 0.6359223300970874, 'overall_f1': 0.6111975116640747, 'overall_accuracy': 0.7663225446428571}
			------------EPOCH 24---------------
Loss:  tensor(21.9260, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.8533, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.4777, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.0186, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.8143, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.1850, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.3812, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.3998, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.5033, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.6604, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.9728, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.5835, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.8015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.5138, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.2191, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.2449, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.2913, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.2895, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.3318, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.5538, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.5109, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.3878, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.6296, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.7831, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.2291, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.4627, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.5735, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.7499, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.0897, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.1319, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.1512, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.7067, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.2497, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.6455, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.0378, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.4764, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.5239, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.3654, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.5867, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.2379, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.8767, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.3472, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.4221, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.5589, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.4339, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.8220, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.7676, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.4583, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.8448, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.4534, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.0745, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.7901, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.3959, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.7467, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.5721, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.3744, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.3475, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.48188405797101447, 'recall': 0.5611814345991561, 'f1': 0.5185185185185185, 'number': 237}, 'P': {'precision': 0.6945169712793734, 'recall': 0.6981627296587927, 'f1': 0.6963350785340315, 'number': 381}, 'overall_precision': 0.6054628224582701, 'overall_recall': 0.6456310679611651, 'overall_f1': 0.6249021143304622, 'overall_accuracy': 0.7686244419642857}
			------------EPOCH 25---------------
Loss:  tensor(20.4803, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.0534, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.7050, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.7047, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.1040, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.1600, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.6684, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.0989, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.3157, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.6373, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.7102, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.9031, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.8466, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.4784, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.6253, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.9200, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.9922, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.6993, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.1869, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.9107, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.5418, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.5255, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.4347, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.2770, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.6835, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.6442, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.6058, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.0011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.1678, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.3539, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.6301, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.3685, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.0768, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.1074, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.4315, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.2055, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.4814, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.7299, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.4609, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.9304, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.7141, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.8893, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.1483, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.1890, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.5514, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.9168, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.5865, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.4384, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.3490, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.6327, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.2731, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.7091, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.7973, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.0396, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.8856, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.7134, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.3344, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.48028673835125446, 'recall': 0.5654008438818565, 'f1': 0.5193798449612402, 'number': 237}, 'P': {'precision': 0.6992084432717678, 'recall': 0.6955380577427821, 'f1': 0.6973684210526315, 'number': 381}, 'overall_precision': 0.6063829787234043, 'overall_recall': 0.6456310679611651, 'overall_f1': 0.6253918495297806, 'overall_accuracy': 0.767578125}
			------------EPOCH 26---------------
Loss:  tensor(18.8700, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.9118, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.5737, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.7618, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.9139, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.9640, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.1432, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.6562, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.0152, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.6456, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.0985, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.0560, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.7324, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.8982, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.9437, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.2497, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.2460, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.9590, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.2111, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.0400, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.6198, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.5326, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.1612, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.0366, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.6966, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.6532, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.3893, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.3597, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.2806, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.8970, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.1900, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.3627, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.6645, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.8097, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.6780, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.7202, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.9262, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.9491, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.2165, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.5549, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.0655, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.1544, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.2392, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.0650, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.0173, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.8506, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.3959, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.8597, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.1296, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.2571, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.5135, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.5635, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.0482, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.6150, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.7336, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.2235, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4910394265232975, 'recall': 0.5780590717299579, 'f1': 0.5310077519379846, 'number': 237}, 'P': {'precision': 0.6992084432717678, 'recall': 0.6955380577427821, 'f1': 0.6973684210526315, 'number': 381}, 'overall_precision': 0.6109422492401215, 'overall_recall': 0.6504854368932039, 'overall_f1': 0.6300940438871473, 'overall_accuracy': 0.7681361607142857}
			------------EPOCH 27---------------
Loss:  tensor(17.6873, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.1029, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.9081, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.8123, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.2866, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.2707, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.6202, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.6426, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.7092, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.5662, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.6276, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.1866, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.5613, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.8984, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.4681, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.3006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.5003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.3916, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.2598, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.3267, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.6248, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.2675, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.6441, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.8956, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.8109, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.0638, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.7606, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.9206, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.5633, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.4486, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.7070, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.0611, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.9519, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.0345, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.3477, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.6698, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.0988, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.7245, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.5768, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.9709, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.5660, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.2214, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.4671, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.4417, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.7664, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.3222, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.5423, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.5861, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.6631, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.5551, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.4691, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.5137, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.7189, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.6793, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.4365, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.8661, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.1111, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4981684981684982, 'recall': 0.5738396624472574, 'f1': 0.5333333333333333, 'number': 237}, 'P': {'precision': 0.6821705426356589, 'recall': 0.6929133858267716, 'f1': 0.6875, 'number': 381}, 'overall_precision': 0.6060606060606061, 'overall_recall': 0.6472491909385113, 'overall_f1': 0.6259780907668231, 'overall_accuracy': 0.7675083705357143}
			------------EPOCH 28---------------
Loss:  tensor(16.5536, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.4407, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.7176, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.1527, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.0962, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.5519, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.2059, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.4248, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.2623, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.3667, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.7366, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.6046, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.0613, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.5014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.9796, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.6382, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.0261, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.3412, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.2988, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.9963, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.3590, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.3761, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.4149, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.0249, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.6246, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.4215, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.4524, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.4721, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.9528, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.2724, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.9227, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.0992, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.1331, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.4718, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.7869, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.8070, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.1110, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.0421, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.9821, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.5246, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.1775, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.6794, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.3501, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.2581, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.1113, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.4969, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.1969, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.0975, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.4810, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.7405, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.1975, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.0862, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.5323, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.8454, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.8067, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.3792, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.2650, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4963768115942029, 'recall': 0.5780590717299579, 'f1': 0.53411306042885, 'number': 237}, 'P': {'precision': 0.6884816753926701, 'recall': 0.6902887139107612, 'f1': 0.6893840104849279, 'number': 381}, 'overall_precision': 0.60790273556231, 'overall_recall': 0.6472491909385113, 'overall_f1': 0.6269592476489029, 'overall_accuracy': 0.7677176339285714}
			------------EPOCH 29---------------
Loss:  tensor(15.4071, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.0355, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.6636, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.1870, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.4708, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.7142, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.1512, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.5277, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.8399, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.6871, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.6975, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.6418, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.9175, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.3673, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.6022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.4288, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.9398, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.9600, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.1889, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.8083, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.7399, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.7792, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.6386, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.2342, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.3004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.9412, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.2466, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.1394, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.7165, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.1846, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.1055, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.3912, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.3064, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.2332, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.8824, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.2583, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.6815, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.2547, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.0721, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.8066, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.7298, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.4784, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.3807, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.1338, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.8133, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.1279, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.0851, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.6530, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.3068, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.1604, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.1969, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.6806, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.4812, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.5126, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.1936, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.6413, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.6824, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4982698961937716, 'recall': 0.6075949367088608, 'f1': 0.5475285171102662, 'number': 237}, 'P': {'precision': 0.6923076923076923, 'recall': 0.6850393700787402, 'f1': 0.6886543535620053, 'number': 381}, 'overall_precision': 0.6081081081081081, 'overall_recall': 0.6553398058252428, 'overall_f1': 0.630841121495327, 'overall_accuracy': 0.7730887276785714}
			------------EPOCH 30---------------
Loss:  tensor(14.6729, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.6660, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.8549, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.9350, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.4822, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.6393, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.6929, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.4357, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.3218, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.9764, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.7675, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.2534, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.2370, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.4748, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.0648, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.4537, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.2768, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.8614, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.5618, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.1751, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.2934, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.8235, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.1492, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.2707, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.4537, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.5914, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.0813, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(80.1452, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(59.7992, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(147.4542, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.7023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(107.4930, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(110.5612, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.4156, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.2375, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(70.6352, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.8282, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.7611, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.8648, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.0977, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.7632, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.5692, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.0136, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.0212, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.0541, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.1498, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(88.6459, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(112.6634, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.4244, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.0217, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.2020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.3987, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(119.8463, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(139.4263, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(169.1385, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.1069, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.2731, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.49640287769784175, 'recall': 0.5822784810126582, 'f1': 0.5359223300970873, 'number': 237}, 'P': {'precision': 0.5938375350140056, 'recall': 0.5564304461942258, 'f1': 0.5745257452574526, 'number': 381}, 'overall_precision': 0.5511811023622047, 'overall_recall': 0.5663430420711975, 'overall_f1': 0.558659217877095, 'overall_accuracy': 0.7458844866071429}


		-------------RUN 4-----------
			------------EPOCH 1---------------
Loss:  tensor(2371.6545, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2283.0513, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2959.7759, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2107.3428, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4349.5977, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3453.1855, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1512.1671, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1655.8240, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1769.7230, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2433.2732, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1090.1483, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2883.4734, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2282.8320, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2506.1738, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1701.9965, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2904.5781, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3380.2910, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2584.6001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2831.2388, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1416.4911, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1478.0791, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1751.5139, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1482.0480, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(999.0873, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2467.9177, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1680.1129, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2939.7820, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3549.6775, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2128.2695, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2762.6953, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1038.6870, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(925.1979, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2548.5371, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2900.9980, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1726.0609, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1786.0785, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1929.6179, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1403.2439, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1686.1095, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1984.3589, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2116.7339, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1395.4180, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2117.1543, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2396.8110, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2106.1125, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2612.1902, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2581.6685, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1238.1333, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1962.8511, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1321.9517, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1112.8804, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(783.1072, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(792.7977, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1980.4855, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1762.2754, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1696.9541, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2591.1702, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.06593406593406594, 'recall': 0.12345679012345678, 'f1': 0.08595988538681949, 'number': 243}, 'P': {'precision': 0.11842105263157894, 'recall': 0.05187319884726225, 'f1': 0.0721442885771543, 'number': 347}, 'overall_precision': 0.07907742998352553, 'overall_recall': 0.08135593220338982, 'overall_f1': 0.08020050125313284, 'overall_accuracy': 0.42459363957597174}
			------------EPOCH 2---------------
Loss:  tensor(1667.8201, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1704.1694, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2143.7627, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1403.3960, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3387.2402, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2567.3401, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1166.2781, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1266.7886, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1394.5793, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1899.0920, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(886.8748, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2299.5977, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1860.4271, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1935.4926, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1331.4302, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2354.5542, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2933.1768, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2216.4822, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2417.4194, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1114.0842, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1101.5148, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1422.3009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1268.6404, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(835.5835, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2024.5042, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1395.7139, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2549.9722, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2853.5586, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1734.5845, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2314.3838, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(928.1824, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(747.9537, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2123.5513, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2378.4854, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1464.6523, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1426.5269, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1546.4780, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1128.7690, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1371.9050, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1783.0198, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1701.2936, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1130.8335, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1734.1646, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2140.6406, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1714.0862, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2143.4956, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2011.1987, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1073.3655, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1778.4583, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1106.8429, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(951.8051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(730.0410, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(690.2814, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1797.1938, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1555.2012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1490.3171, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2468.6272, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.23711340206185566, 'recall': 0.2839506172839506, 'f1': 0.25842696629213485, 'number': 243}, 'P': {'precision': 0.21749408983451538, 'recall': 0.26512968299711814, 'f1': 0.23896103896103896, 'number': 347}, 'overall_precision': 0.22549019607843138, 'overall_recall': 0.27288135593220336, 'overall_f1': 0.24693251533742328, 'overall_accuracy': 0.620565371024735}
			------------EPOCH 3---------------
Loss:  tensor(1429.9180, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1560.7611, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1854.4819, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1131.6267, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3141.7913, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2207.6709, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1037.2378, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1090.5552, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1176.9723, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1649.5645, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(805.4170, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2013.8010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1585.8928, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1590.9213, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1069.6304, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1983.2351, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2392.7739, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1921.4714, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2108.2964, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(970.1384, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(855.1138, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1253.8135, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1169.8330, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(673.5394, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1634.1091, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1115.9656, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2117.0161, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2334.5312, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1389.7249, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1986.2053, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(834.0745, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(594.0374, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1784.9773, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1994.3707, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1246.0856, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1173.1621, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1227.0662, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(882.4084, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1122.9783, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1531.0386, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1328.3884, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(879.8439, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1460.2246, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1911.5083, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1355.0482, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1773.0304, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1629.8408, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(958.9457, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1435.3708, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(887.3535, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(748.1816, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(574.8622, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(548.6632, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1151.1713, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(984.5228, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(936.7166, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2059.6650, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.24834437086092714, 'recall': 0.30864197530864196, 'f1': 0.27522935779816515, 'number': 243}, 'P': {'precision': 0.26644736842105265, 'recall': 0.2334293948126801, 'f1': 0.2488479262672811, 'number': 347}, 'overall_precision': 0.25742574257425743, 'overall_recall': 0.26440677966101694, 'overall_f1': 0.26086956521739124, 'overall_accuracy': 0.6032508833922261}
			------------EPOCH 4---------------
Loss:  tensor(1072.5580, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1423.6624, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1523.3149, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(987.7830, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2768.4067, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1715.8101, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(804.9652, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(840.2279, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(940.4717, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1404.3767, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(620.5668, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1596.8856, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1246.7164, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1306.5759, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(766.3971, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1774.0275, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2181.3267, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1714.6915, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2003.4281, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(838.1832, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(565.4335, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(963.2082, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1052.1233, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(572.9141, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1214.3768, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(693.1781, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1625.6061, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1765.7539, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1037.5752, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1647.6869, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(690.6367, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(441.0385, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1364.3309, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1640.5785, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(894.5793, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(923.6786, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(934.6367, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(656.3918, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(864.1709, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1198.0959, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1042.2915, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(704.8853, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1139.6943, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1634.8218, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(960.2708, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1382.9932, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1296.7372, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(878.2586, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1165.1591, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(721.3651, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(586.1442, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(434.0112, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(414.8552, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(758.7783, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(631.3976, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(542.3199, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1736.0901, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3888888888888889, 'recall': 0.43209876543209874, 'f1': 0.40935672514619886, 'number': 243}, 'P': {'precision': 0.29069767441860467, 'recall': 0.36023054755043227, 'f1': 0.3217503217503218, 'number': 347}, 'overall_precision': 0.32857142857142857, 'overall_recall': 0.3898305084745763, 'overall_f1': 0.3565891472868217, 'overall_accuracy': 0.6920848056537102}
			------------EPOCH 5---------------
Loss:  tensor(831.2305, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1032.1802, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1254.0538, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(654.3458, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2215.5686, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1243.9020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(724.9905, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(836.9382, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(748.6386, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1454.2881, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(591.5829, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1382.6437, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(976.8995, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1056.3414, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(615.5118, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1457.0221, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1224.2490, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1149.7865, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1634.1796, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(677.7820, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(353.2925, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(682.0715, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(808.8441, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(365.0033, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(901.8033, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(439.9822, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1256.2180, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1234.0862, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(710.0980, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1327.3583, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(698.0842, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(419.0654, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1143.3198, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1321.1758, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(717.8242, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(735.6960, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(542.1051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(454.4908, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(661.3307, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(940.3856, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(705.7006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(561.6957, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(860.6337, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1308.4766, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(791.0643, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1192.7207, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1087.7426, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(622.5665, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(775.4711, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(530.2393, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(335.4866, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(320.0089, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(304.2943, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(495.5067, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(336.2528, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(316.7451, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1320.3440, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4230769230769231, 'recall': 0.22633744855967078, 'f1': 0.2949061662198391, 'number': 243}, 'P': {'precision': 0.4331983805668016, 'recall': 0.6167146974063401, 'f1': 0.5089179548156955, 'number': 347}, 'overall_precision': 0.4310897435897436, 'overall_recall': 0.4559322033898305, 'overall_f1': 0.443163097199341, 'overall_accuracy': 0.6911660777385159}
			------------EPOCH 6---------------
Loss:  tensor(690.9833, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(991.2181, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1101.4404, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(555.3842, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2052.3457, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1058.4021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(711.7930, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(865.0925, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(635.2540, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1477.8577, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(426.8521, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(841.4893, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(645.2474, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1171.1515, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(400.9608, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1092.9680, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1147.0349, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1177.7771, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1446.0303, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(782.6790, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(295.9797, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(674.3698, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(683.1681, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(293.3138, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(799.1984, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(367.2973, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(804.3088, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(984.9948, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(659.9136, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1053.4890, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(609.7684, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(305.8645, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1091.2994, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1188.1333, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(747.8972, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(688.6428, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(488.7103, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(453.1524, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(711.5077, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1186.6812, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(861.9280, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(525.0991, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1174.7515, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1921.2015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1005.7168, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1153.7382, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(641.8146, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(665.9813, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1046.4517, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(360.1263, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(330.3311, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(274.1449, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(326.5766, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(397.2519, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(338.9815, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(310.2566, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1137.3000, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.2479108635097493, 'recall': 0.3662551440329218, 'f1': 0.2956810631229236, 'number': 243}, 'P': {'precision': 0.20647773279352227, 'recall': 0.14697406340057637, 'f1': 0.17171717171717174, 'number': 347}, 'overall_precision': 0.23102310231023102, 'overall_recall': 0.23728813559322035, 'overall_f1': 0.23411371237458195, 'overall_accuracy': 0.6257950530035336}
			------------EPOCH 7---------------
Loss:  tensor(842.8503, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(991.6220, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1027.7573, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(478.9327, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2129.3572, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(925.2338, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(552.1911, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(705.1847, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(569.6568, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1102.2163, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(427.7130, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(690.9962, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(567.9197, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(991.8102, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(375.9378, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(787.9014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(909.4902, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1134.2317, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1175.7589, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(732.7233, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(291.8528, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(528.0610, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(421.0461, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(303.9739, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(545.4072, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(283.5009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(566.3294, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(710.9598, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(502.7069, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(918.4681, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(436.6516, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(211.4683, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(788.8390, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(879.5631, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(575.4622, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(556.4935, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(377.6474, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(298.9201, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(461.4400, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(774.2750, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(492.9676, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(410.4668, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(639.4742, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(827.8201, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(496.7042, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(684.5162, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(417.4826, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(428.4258, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(778.5770, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(307.1281, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(248.0076, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(198.2900, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(287.8113, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(351.4399, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(261.8580, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(223.2010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1099.7136, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.38073394495412843, 'recall': 0.34156378600823045, 'f1': 0.3600867678958785, 'number': 243}, 'P': {'precision': 0.42054263565891475, 'recall': 0.6253602305475504, 'f1': 0.5028968713789108, 'number': 347}, 'overall_precision': 0.4087193460490463, 'overall_recall': 0.5084745762711864, 'overall_f1': 0.4531722054380664, 'overall_accuracy': 0.7016254416961131}
			------------EPOCH 8---------------
Loss:  tensor(422.0967, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(970.7667, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(822.5438, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(339.3553, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1557.9197, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(558.9294, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(534.7585, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(685.4874, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(442.5462, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(835.3510, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(277.7756, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(429.1858, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(356.1593, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(723.2047, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(280.6248, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(501.8831, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(468.3563, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(526.1488, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(655.2291, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(431.4680, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(209.2625, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(607.0557, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(452.8742, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(265.0860, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(903.1476, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(459.8715, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(749.0042, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1034.6907, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(540.8059, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1190.5717, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(514.7172, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(189.7995, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(596.1312, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(608.1660, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(312.2175, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(355.9976, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(193.0565, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(186.4052, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(338.5214, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(452.5792, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(273.3353, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(262.7104, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(435.5263, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(653.5660, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(351.8155, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(481.1193, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(266.0458, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(296.5986, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(623.8191, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(160.1444, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(139.7884, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(146.7222, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(278.3567, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(251.9728, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(219.7410, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(176.8668, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(901.2953, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.43873517786561267, 'recall': 0.4567901234567901, 'f1': 0.4475806451612903, 'number': 243}, 'P': {'precision': 0.48532289628180036, 'recall': 0.7146974063400576, 'f1': 0.578088578088578, 'number': 347}, 'overall_precision': 0.4698952879581152, 'overall_recall': 0.6084745762711864, 'overall_f1': 0.5302806499261448, 'overall_accuracy': 0.6831802120141343}
			------------EPOCH 9---------------
Loss:  tensor(343.3946, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(776.7652, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(647.8654, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(331.1690, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1339.2808, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(543.8217, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(628.2003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(717.5535, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(505.9202, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(706.3304, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(234.3627, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(916.9900, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(480.9058, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(756.3098, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(183.8379, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(737.6345, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(607.9358, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(564.6437, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(547.6798, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(396.2249, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(111.1060, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(242.9203, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(231.5040, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(134.6375, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(354.7796, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(136.8205, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(384.3547, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(465.1978, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(378.2642, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(712.3335, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(338.8117, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(140.9913, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(743.9072, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(711.8135, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(359.6473, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(462.7139, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(579.9954, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(553.8264, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(597.4604, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(776.8439, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(784.1530, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(709.1892, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(652.9635, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(833.2722, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(311.4265, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(421.8267, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(213.4676, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(349.0887, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(280.3070, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(179.4497, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(116.9584, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(156.3944, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(175.9601, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(138.5251, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(101.7601, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(95.9144, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(400.8098, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3382352941176471, 'recall': 0.3786008230452675, 'f1': 0.35728155339805834, 'number': 243}, 'P': {'precision': 0.4144329896907217, 'recall': 0.579250720461095, 'f1': 0.48317307692307687, 'number': 347}, 'overall_precision': 0.3870541611624835, 'overall_recall': 0.4966101694915254, 'overall_f1': 0.43504083147735706, 'overall_accuracy': 0.705583038869258}
			------------EPOCH 10---------------
Loss:  tensor(244.3948, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(331.0883, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(301.5919, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(114.1336, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(771.1818, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(208.1850, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(271.4203, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(400.8576, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(472.5728, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(487.9764, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(137.1354, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(323.5671, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(292.4201, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(880.1964, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(316.6093, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(751.8221, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(644.6366, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(551.6879, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1000.1553, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(221.8194, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(119.7161, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(439.6042, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(496.4732, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(329.0465, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(325.5602, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(214.5644, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(413.5983, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(530.3855, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(429.1620, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(463.4186, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(283.5154, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(105.7551, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(475.2335, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(365.5138, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(253.8661, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(220.0317, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(118.5839, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(108.3408, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(307.4301, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(366.5519, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(161.5078, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(218.6787, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(334.9841, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(427.7708, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(307.3490, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(307.4164, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(300.5725, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(331.1945, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(320.6261, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(179.2754, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(103.0635, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(202.2855, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(118.2316, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(160.6149, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(127.0536, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(109.5093, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(345.1486, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5102040816326531, 'recall': 0.411522633744856, 'f1': 0.45558086560364464, 'number': 243}, 'P': {'precision': 0.5529100529100529, 'recall': 0.6023054755043228, 'f1': 0.576551724137931, 'number': 347}, 'overall_precision': 0.5383275261324042, 'overall_recall': 0.523728813559322, 'overall_f1': 0.5309278350515464, 'overall_accuracy': 0.7265724381625441}
			------------EPOCH 11---------------
Loss:  tensor(167.7308, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(664.1470, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(364.8291, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(170.2457, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(798.5751, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(477.4716, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(403.6370, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(446.1167, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(233.0975, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(357.8316, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(197.0204, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(240.2338, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(178.9205, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(597.2834, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(153.3143, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(278.8928, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(173.2017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(159.3913, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(215.6572, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(229.7583, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(62.3331, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(148.2113, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(207.0396, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(97.8118, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(213.7839, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(95.4843, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(245.1959, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(336.6455, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(214.0232, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(491.3433, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(162.3353, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(77.9480, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(288.6373, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(215.8604, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(211.2175, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(190.4227, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(80.8830, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(80.6562, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(168.2436, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(231.0477, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(153.4075, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(212.7090, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(349.1769, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(570.6091, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(287.6954, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(374.2499, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(171.9826, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(178.0748, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(499.0905, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(130.0632, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(111.0782, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(117.6095, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(174.3989, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(125.4901, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(96.1233, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(73.8449, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(353.7476, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.41475826972010177, 'recall': 0.6707818930041153, 'f1': 0.5125786163522011, 'number': 243}, 'P': {'precision': 0.602112676056338, 'recall': 0.49279538904899134, 'f1': 0.5419968304278922, 'number': 347}, 'overall_precision': 0.4933530280649926, 'overall_recall': 0.5661016949152542, 'overall_f1': 0.527229676400947, 'overall_accuracy': 0.6925088339222615}
			------------EPOCH 12---------------
Loss:  tensor(151.8997, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(432.8065, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(317.2151, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(139.2341, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(387.2488, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(119.8032, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(167.5904, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(105.3373, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(128.9937, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(171.6624, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.0781, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(105.6799, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(100.7559, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(523.5670, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(73.1386, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(151.4339, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(137.6573, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(173.3897, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(298.2933, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(222.7130, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(224.9497, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(212.6412, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(260.3593, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(245.2589, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(241.7606, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(138.2045, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(339.0197, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(431.0646, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(242.5774, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(372.1727, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(282.7839, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(394.8472, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(268.6571, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(186.1060, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(68.3532, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(147.1510, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.5740, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(89.0540, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(209.3438, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(161.4763, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(103.5232, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(135.3059, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(143.6564, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(239.5048, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(180.8725, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(203.7858, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(142.3442, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(118.9467, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(165.3087, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(159.4684, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(77.4926, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(145.5851, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(169.8273, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(93.5260, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(126.5529, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(103.8342, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(515.2711, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.30755064456721914, 'recall': 0.6872427983539094, 'f1': 0.4249363867684478, 'number': 243}, 'P': {'precision': 0.5833333333333334, 'recall': 0.2017291066282421, 'f1': 0.2997858672376874, 'number': 347}, 'overall_precision': 0.3574660633484163, 'overall_recall': 0.4016949152542373, 'overall_f1': 0.37829209896249, 'overall_accuracy': 0.5621201413427562}
			------------EPOCH 13---------------
Loss:  tensor(250.5112, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(588.4194, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(547.8750, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(185.8368, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(618.8626, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(253.9177, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(413.3558, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(268.5630, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(151.6745, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(163.4791, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(96.6474, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(168.3505, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(71.5236, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(325.5696, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(62.4876, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(109.3748, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(100.7127, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(72.0627, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(138.6608, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(114.2129, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.2389, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(106.0361, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(123.6390, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(103.1739, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(329.7170, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(133.9700, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(287.3053, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(452.5970, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(456.0655, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(972.9938, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(705.5336, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(883.9765, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2070.3193, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1406.9996, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(216.3036, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1185.9568, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(242.9542, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(405.9863, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(526.2770, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(596.6067, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(215.3961, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(109.1273, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(248.5693, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(451.0049, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(203.6877, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(137.8823, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(119.7705, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(158.4284, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(146.2014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(122.2496, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.3360, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(107.4986, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(120.3763, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(83.7893, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(61.3084, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(66.4375, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(343.5651, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.30033557046979864, 'recall': 0.7366255144032922, 'f1': 0.42669845053635275, 'number': 243}, 'P': {'precision': 0.72, 'recall': 0.15561959654178675, 'f1': 0.2559241706161138, 'number': 347}, 'overall_precision': 0.3472429210134128, 'overall_recall': 0.3949152542372881, 'overall_f1': 0.36954797779540044, 'overall_accuracy': 0.5240282685512367}
			------------EPOCH 14---------------
Loss:  tensor(200.5845, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(472.7834, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(808.4056, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(334.5000, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1906.0510, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(556.3046, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(380.0614, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(389.6810, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(622.0244, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1027.3130, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(419.0103, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1219.6580, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(670.0852, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1111.0414, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(235.2270, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1104.9777, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1180.6216, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(804.4990, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(645.5983, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(204.9129, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.1230, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(128.3483, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(193.6748, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(87.5405, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(138.6268, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(97.9127, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(182.0257, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(289.2556, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(168.0618, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(244.6628, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(98.8043, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(86.9496, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(322.4865, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(291.7351, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(126.2950, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(277.5107, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(171.2246, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(188.0437, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(534.3607, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(359.1313, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(717.0758, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(296.7952, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(962.1484, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1046.7736, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1043.3076, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1014.2509, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(798.5797, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1227.3718, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(570.0690, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(424.2677, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(364.4276, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(226.1968, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(189.1631, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(414.7610, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(233.8166, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(315.2097, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(611.3851, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5371900826446281, 'recall': 0.5349794238683128, 'f1': 0.5360824742268041, 'number': 243}, 'P': {'precision': 0.58679706601467, 'recall': 0.69164265129683, 'f1': 0.6349206349206349, 'number': 347}, 'overall_precision': 0.5683563748079877, 'overall_recall': 0.6271186440677966, 'overall_f1': 0.5962933118452861, 'overall_accuracy': 0.7603533568904594}
			------------EPOCH 15---------------
Loss:  tensor(281.4300, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(313.8327, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(323.0018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(101.9257, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(444.4213, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(275.2134, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(185.3463, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(173.7725, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(150.1938, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(244.9826, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(106.0088, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(138.7867, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(141.9825, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(319.5635, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(104.3325, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(203.6789, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(324.9539, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(216.3393, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(349.9755, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(125.5586, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.0284, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(123.2205, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(186.5696, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(95.7015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(363.6711, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(198.7476, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(318.7319, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(683.1343, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(498.1090, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(563.5488, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(153.2340, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(71.4403, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(395.7702, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(316.5168, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(348.5913, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(219.1929, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(176.0262, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(130.1483, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(282.9595, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(237.8601, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(115.8922, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(147.7774, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(135.3824, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(258.9022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(119.2883, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(149.2988, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(143.6856, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(126.6280, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(77.9123, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.1088, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.5467, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(77.6779, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.3998, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(112.0314, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.6430, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.8078, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(97.3605, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4391025641025641, 'recall': 0.5637860082304527, 'f1': 0.4936936936936937, 'number': 243}, 'P': {'precision': 0.6125356125356125, 'recall': 0.6195965417867435, 'f1': 0.6160458452722063, 'number': 347}, 'overall_precision': 0.530920060331825, 'overall_recall': 0.5966101694915255, 'overall_f1': 0.5618515562649641, 'overall_accuracy': 0.7397173144876326}
			------------EPOCH 16---------------
Loss:  tensor(89.3849, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(141.5020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(113.7029, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.8934, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(250.2065, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(101.8304, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(114.5921, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(74.8854, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(111.7993, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(166.9721, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.3829, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(99.5640, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(73.8672, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(235.1720, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(106.9359, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(94.7778, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(149.3873, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(68.0702, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(112.3497, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(88.4447, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.6951, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(67.4890, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(98.1462, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(64.7145, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(73.2818, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.0951, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(126.4424, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(212.0709, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(164.0853, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(117.3277, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.6954, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.6217, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(115.9412, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(74.1771, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.0876, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(79.4578, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.1062, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.6412, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(112.6939, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(73.9933, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.5578, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(72.6389, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(152.5246, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(194.6821, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(89.1043, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(69.6783, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.1450, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(107.5586, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(90.3840, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.7271, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.0676, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.6140, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.0814, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(100.2383, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.5976, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.9177, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(67.2879, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5441176470588235, 'recall': 0.6090534979423868, 'f1': 0.574757281553398, 'number': 243}, 'P': {'precision': 0.609254498714653, 'recall': 0.6829971181556196, 'f1': 0.6440217391304348, 'number': 347}, 'overall_precision': 0.5824508320726173, 'overall_recall': 0.652542372881356, 'overall_f1': 0.6155075939248602, 'overall_accuracy': 0.7631802120141343}
			------------EPOCH 17---------------
Loss:  tensor(55.2746, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(84.7475, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(90.4041, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.4226, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(164.2325, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.4553, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(98.5550, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.6059, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(70.2816, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(85.8387, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.6783, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(65.2814, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.8626, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(172.7776, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.7177, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.2323, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(78.6541, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.9704, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(90.2054, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.1510, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.4932, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.6609, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(87.7997, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.0665, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.8082, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.4808, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(110.7450, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(176.4103, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(128.6792, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(78.6364, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.0560, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.2946, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(77.4471, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.2685, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.2958, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.0298, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.1868, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.9730, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(78.9981, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(64.7363, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.7001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.9705, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.4821, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.9861, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(59.9514, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.9089, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.9709, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(64.0796, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.5987, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.3433, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.4428, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.1311, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.8074, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.0505, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.7679, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.4338, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.1476, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4902597402597403, 'recall': 0.6213991769547325, 'f1': 0.5480943738656988, 'number': 243}, 'P': {'precision': 0.6285714285714286, 'recall': 0.6340057636887608, 'f1': 0.6312769010043041, 'number': 347}, 'overall_precision': 0.5638297872340425, 'overall_recall': 0.6288135593220339, 'overall_f1': 0.594551282051282, 'overall_accuracy': 0.747208480565371}
			------------EPOCH 18---------------
Loss:  tensor(33.7923, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.7394, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(64.4876, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.9852, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(137.6058, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.3547, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(61.3512, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.2143, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.9457, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.7050, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.7113, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.5068, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.0037, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(112.2366, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.1327, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.5495, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(65.4302, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.6651, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(79.6535, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.8180, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.6843, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.1305, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.5847, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.7495, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.2158, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.0414, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(95.6730, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(150.7448, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(87.7385, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(69.8394, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.9899, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.7622, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.0802, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.2090, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.7679, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.5895, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.5407, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.5924, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.6323, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.6044, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.2462, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.0907, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.4298, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.3427, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.9726, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.3289, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.9919, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.9887, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.3725, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.1124, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.1028, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.1551, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.2170, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.2044, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.7059, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.6592, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.5187, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5223367697594502, 'recall': 0.6255144032921811, 'f1': 0.5692883895131086, 'number': 243}, 'P': {'precision': 0.635593220338983, 'recall': 0.6484149855907781, 'f1': 0.6419400855920114, 'number': 347}, 'overall_precision': 0.5844961240310077, 'overall_recall': 0.6389830508474577, 'overall_f1': 0.6105263157894737, 'overall_accuracy': 0.7498939929328622}
			------------EPOCH 19---------------
Loss:  tensor(31.5256, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.6288, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(61.1872, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.9299, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(148.4925, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.8322, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.9468, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.1592, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.4294, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.2736, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.8692, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.1635, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.0257, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(87.8123, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.3046, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.4646, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.6513, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.7436, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(74.1088, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.6311, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.2835, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.1366, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.3082, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.2367, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.5757, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.3243, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(86.4797, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(138.7360, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(78.4469, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(65.7366, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.6602, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.8900, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.6042, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.8767, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.7029, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.6689, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.7428, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.7829, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.7360, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.8221, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.6016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.8730, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.8720, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.8385, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.5878, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.3220, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.5977, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.1930, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.3794, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.3870, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.4009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.3401, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.7402, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.6104, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.1328, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.1991, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.7305, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5209790209790209, 'recall': 0.6131687242798354, 'f1': 0.5633270321361058, 'number': 243}, 'P': {'precision': 0.6330532212885154, 'recall': 0.6512968299711815, 'f1': 0.6420454545454546, 'number': 347}, 'overall_precision': 0.583203732503888, 'overall_recall': 0.635593220338983, 'overall_f1': 0.6082725060827251, 'overall_accuracy': 0.7498939929328622}
			------------EPOCH 20---------------
Loss:  tensor(28.9109, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.2364, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.6738, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.9233, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(125.6362, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.4877, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.4317, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.0671, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.0197, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.7696, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.2331, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.0208, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.1121, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(71.9835, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.6148, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.6419, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.0355, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.1804, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(67.0722, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.8634, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.2789, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.9121, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.7902, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.7537, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.6025, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.1611, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(75.5561, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(119.2081, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(74.8042, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.0387, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.1362, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.9209, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.9463, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.6717, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.2087, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.7468, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.6462, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.6759, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(129.0068, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(61.3398, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(108.7773, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(96.1001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(87.1316, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(112.5761, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(77.3213, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(122.1854, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(205.6352, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(90.7981, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(73.9975, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.6776, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.7561, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(72.8474, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(103.5463, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(59.1096, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.9968, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.6606, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(150.6008, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.37222222222222223, 'recall': 0.551440329218107, 'f1': 0.4444444444444445, 'number': 243}, 'P': {'precision': 0.5217391304347826, 'recall': 0.5187319884726225, 'f1': 0.5202312138728324, 'number': 347}, 'overall_precision': 0.4453900709219858, 'overall_recall': 0.5322033898305085, 'overall_f1': 0.4849420849420849, 'overall_accuracy': 0.6973851590106007}
			------------EPOCH 21---------------
Loss:  tensor(89.8811, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(171.1133, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(99.2575, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.5426, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(238.2674, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(61.8080, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(65.7276, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.9931, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(82.5149, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(179.9336, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.3819, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(191.1159, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(116.5220, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(195.4545, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.8005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(193.7169, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(183.6614, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.4905, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(126.2131, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(73.0381, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.2067, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(90.6701, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(103.3476, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.3074, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(103.5315, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.1706, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(115.7515, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(210.2520, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(118.6043, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(145.5628, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.6248, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.9178, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(155.5508, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(171.8365, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(68.7705, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.7252, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.0976, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.8145, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(139.3770, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(111.9405, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(99.0173, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(81.8835, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(111.1195, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(94.2871, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.9372, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(61.9068, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(94.0437, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.4757, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.8789, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.7589, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.5586, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.4671, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.1927, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.3279, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.8261, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.3046, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.3694, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.44370860927152317, 'recall': 0.551440329218107, 'f1': 0.4917431192660551, 'number': 243}, 'P': {'precision': 0.6170212765957447, 'recall': 0.6685878962536023, 'f1': 0.6417704011065007, 'number': 347}, 'overall_precision': 0.5398230088495575, 'overall_recall': 0.6203389830508474, 'overall_f1': 0.5772870662460567, 'overall_accuracy': 0.7253003533568905}
			------------EPOCH 22---------------
Loss:  tensor(50.6577, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(145.9543, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(78.8769, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.9273, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(187.2407, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(71.8192, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.1785, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.3184, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.9057, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.5243, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.4285, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.9086, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.5951, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.1001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.7800, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.0652, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.0441, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.9320, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.9331, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.0501, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.9662, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.1632, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.0968, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.9760, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.3876, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.3917, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(86.2290, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(146.0924, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(92.9709, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.5646, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.4009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.5583, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.6071, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.8959, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.6262, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.8989, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.9331, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.5422, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(67.1018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.2633, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.3217, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.7731, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.3538, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.0877, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.3066, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.0118, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.5493, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.2238, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.5029, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.2891, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.3437, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.7533, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.5898, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.7599, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.9957, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.0093, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.6233, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4965986394557823, 'recall': 0.6008230452674898, 'f1': 0.5437616387337058, 'number': 243}, 'P': {'precision': 0.5850144092219021, 'recall': 0.5850144092219021, 'f1': 0.5850144092219021, 'number': 347}, 'overall_precision': 0.5444617784711389, 'overall_recall': 0.5915254237288136, 'overall_f1': 0.5670186839967506, 'overall_accuracy': 0.732226148409894}
			------------EPOCH 23---------------
Loss:  tensor(15.8812, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.6793, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.8128, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.8758, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(152.2132, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.8798, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.9117, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.8265, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.9887, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.1076, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.9196, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.9026, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.2566, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.7424, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.4864, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.1939, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.0536, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.1771, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.2998, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.7169, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.2667, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.2937, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.6002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.7769, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.9747, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(69.6554, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(110.1130, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.1985, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.7535, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.6394, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.0162, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.2941, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.5023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.4755, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.0778, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.0749, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.6855, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.8474, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.7036, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.5973, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.9223, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.6069, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.7115, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.7988, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.0398, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.5657, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.4535, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.2362, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.3499, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.4031, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.2754, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.5325, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.0945, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.6494, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.3399, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.6891, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.49504950495049505, 'recall': 0.6172839506172839, 'f1': 0.5494505494505494, 'number': 243}, 'P': {'precision': 0.6296296296296297, 'recall': 0.6368876080691642, 'f1': 0.6332378223495702, 'number': 347}, 'overall_precision': 0.5672782874617737, 'overall_recall': 0.6288135593220339, 'overall_f1': 0.5964630225080385, 'overall_accuracy': 0.7340636042402827}
			------------EPOCH 24---------------
Loss:  tensor(5.7395, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.1268, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.6214, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.5162, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(107.4026, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.0254, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.7246, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.8769, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.1352, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.5237, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.8018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.8359, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.4610, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.5743, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.8059, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.3869, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.6344, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.8190, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.9725, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.9669, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.2697, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.1627, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.6653, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.1241, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.8291, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.4730, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.5354, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(85.0143, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.0446, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.5759, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.3898, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.1856, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.4903, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.4651, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.4743, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.6205, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.8186, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.9347, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.3704, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.4629, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.6533, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.9005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.7973, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.7381, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.5154, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.3404, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.6878, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.1326, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.2887, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.4802, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.1969, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.7691, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.9248, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.3751, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.5271, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.0540, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.3140, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5033112582781457, 'recall': 0.6255144032921811, 'f1': 0.5577981651376146, 'number': 243}, 'P': {'precision': 0.6314285714285715, 'recall': 0.6368876080691642, 'f1': 0.6341463414634145, 'number': 347}, 'overall_precision': 0.5720858895705522, 'overall_recall': 0.6322033898305085, 'overall_f1': 0.6006441223832528, 'overall_accuracy': 0.7345583038869258}
			------------EPOCH 25---------------
Loss:  tensor(4.0613, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.3905, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.1040, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.1066, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(87.3867, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.9003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.3703, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.8854, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.0884, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.1639, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5752, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.0891, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.9001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.9735, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.0091, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.9377, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.2268, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.8063, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.0942, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.0487, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.0654, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.5587, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.3671, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.5426, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.6117, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.5308, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.6942, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(77.6700, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.7633, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.7644, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.3763, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.9574, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.5259, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.2085, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.8190, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.0057, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.0870, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.2383, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.6085, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.8470, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.6971, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.6392, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.4971, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.3918, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.4944, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.4993, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.2419, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.6880, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.6773, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.4522, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.0639, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.2902, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.8285, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.3023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.0078, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.3059, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.4518, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5051194539249146, 'recall': 0.6090534979423868, 'f1': 0.5522388059701493, 'number': 243}, 'P': {'precision': 0.625, 'recall': 0.6340057636887608, 'f1': 0.6294706723891274, 'number': 347}, 'overall_precision': 0.5705426356589147, 'overall_recall': 0.6237288135593221, 'overall_f1': 0.5959514170040485, 'overall_accuracy': 0.7311660777385159}
			------------EPOCH 26---------------
Loss:  tensor(3.2566, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.8103, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.4997, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.3158, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(79.8062, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.7247, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.5187, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.6387, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.8753, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.2873, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.1804, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.3488, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.4256, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.0717, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.0786, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.8712, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.1414, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.1943, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.1769, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.5743, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.3844, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.3207, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.9803, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.3541, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.0722, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.6638, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.5361, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.4108, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.5904, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.6422, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.7869, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.4519, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.0948, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.0027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.1333, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.3918, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.4099, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.1909, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.4995, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.9592, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.8970, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.0114, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.7966, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.4956, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.0980, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.4925, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.4925, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.4743, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.8858, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.3400, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5375, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.9175, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5147, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.9932, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.3097, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.6426, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.6195, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5, 'recall': 0.6090534979423868, 'f1': 0.549165120593692, 'number': 243}, 'P': {'precision': 0.6228571428571429, 'recall': 0.6282420749279539, 'f1': 0.6255380200860833, 'number': 347}, 'overall_precision': 0.56656346749226, 'overall_recall': 0.6203389830508474, 'overall_f1': 0.5922330097087378, 'overall_accuracy': 0.7260070671378092}
			------------EPOCH 27---------------
Loss:  tensor(2.8979, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.5737, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.5740, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.7392, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(71.9792, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.2919, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.9291, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.9172, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.3339, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.1946, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.8460, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.5525, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.8919, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.3179, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.4550, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.2764, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.5164, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.6324, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.5828, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.5432, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.0454, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5681, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.5044, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.4236, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.9023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.3858, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.3095, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.7496, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.0913, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.9321, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.3769, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.0690, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.9941, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.1952, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.7103, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.6598, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.9609, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.1113, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.9066, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.1518, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.3409, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.0155, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.7536, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.9964, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.8989, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.6852, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.1142, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.7227, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.2289, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.0036, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.9924, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.9040, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.9816, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.7409, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.0286, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.3111, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.1640, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4966216216216216, 'recall': 0.6049382716049383, 'f1': 0.5454545454545455, 'number': 243}, 'P': {'precision': 0.6257142857142857, 'recall': 0.6311239193083573, 'f1': 0.6284074605451937, 'number': 347}, 'overall_precision': 0.56656346749226, 'overall_recall': 0.6203389830508474, 'overall_f1': 0.5922330097087378, 'overall_accuracy': 0.726148409893993}
			------------EPOCH 28---------------
Loss:  tensor(2.4179, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.0675, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.7583, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.3646, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.6652, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.4491, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.4457, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.4672, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.8482, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.5722, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6208, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.0611, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5192, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.8223, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.1432, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.8803, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.6901, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.3306, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.3790, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.5622, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.8371, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2092, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.7518, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.9951, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.3610, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.0627, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.6454, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.9001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.0476, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.2955, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.0081, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6864, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.2101, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.6123, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2922, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.0805, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5757, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.4523, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.5357, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.6407, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.8995, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.2767, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.0231, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.9991, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.3576, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.9944, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.8451, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.1997, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.7818, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.7878, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6702, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.0070, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6541, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.8333, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.8369, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.0859, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.8355, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4915254237288136, 'recall': 0.5967078189300411, 'f1': 0.5390334572490707, 'number': 243}, 'P': {'precision': 0.6257142857142857, 'recall': 0.6311239193083573, 'f1': 0.6284074605451937, 'number': 347}, 'overall_precision': 0.5643410852713179, 'overall_recall': 0.6169491525423729, 'overall_f1': 0.5894736842105264, 'overall_accuracy': 0.7268551236749117}
			------------EPOCH 29---------------
Loss:  tensor(2.0296, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.7171, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.2039, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.0740, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.8303, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.7158, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.1342, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.1556, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.0787, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.1305, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4520, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.7258, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2211, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.4013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.8847, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5898, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.2767, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.0920, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.6621, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.2368, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6692, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.9630, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.3072, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.3610, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.9888, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.8888, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.3458, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.7440, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.1264, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.9984, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.7611, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4898, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.6704, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.2220, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.0318, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.7076, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2941, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.9061, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.3768, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.1996, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5867, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.8052, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.4597, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.3681, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.3954, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.3727, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.6511, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.7733, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.4730, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5991, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4574, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.2446, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4396, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.1231, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6642, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.8925, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5747, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.49337748344370863, 'recall': 0.6131687242798354, 'f1': 0.5467889908256881, 'number': 243}, 'P': {'precision': 0.6282420749279539, 'recall': 0.6282420749279539, 'f1': 0.6282420749279539, 'number': 347}, 'overall_precision': 0.5654853620955316, 'overall_recall': 0.6220338983050847, 'overall_f1': 0.5924132364810332, 'overall_accuracy': 0.7237455830388693}
			------------EPOCH 30---------------
Loss:  tensor(1.7995, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.3721, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.8557, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.8535, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.9229, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.1142, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.9220, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.9245, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.4570, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.8190, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3061, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.4575, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.9853, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.9309, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6807, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.3433, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.3873, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.9015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.1890, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.1659, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5196, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.7760, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.0227, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.9718, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.6565, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.7036, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.1686, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.7846, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.3597, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.5149, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5651, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2986, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.2202, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.8602, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.8032, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.3932, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.0455, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.4662, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.0940, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.8216, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.3197, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.4239, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.1472, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.9903, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.6250, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.3343, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.4088, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.4958, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.1953, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4318, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2900, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.5284, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2880, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.3982, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4739, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6942, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.3210, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4883720930232558, 'recall': 0.6049382716049383, 'f1': 0.5404411764705882, 'number': 243}, 'P': {'precision': 0.6282420749279539, 'recall': 0.6282420749279539, 'f1': 0.6282420749279539, 'number': 347}, 'overall_precision': 0.5632716049382716, 'overall_recall': 0.6186440677966102, 'overall_f1': 0.5896607431340873, 'overall_accuracy': 0.7238162544169612}


		-------------RUN 5-----------
			------------EPOCH 1---------------
Loss:  tensor(1525.7628, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1639.5638, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1954.5291, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2619.1199, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1070.0645, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2802.9951, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2256.5889, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2060.3367, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1694.5133, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3040.1472, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1945.2822, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1242.5535, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(919.4644, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2226.8418, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1293.3560, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2081.0645, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1212.5240, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2191.9243, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3530.5469, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2773.9951, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2577.1401, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2317.1021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1058.4445, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1157.9617, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2088.0454, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2715.4177, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1065.6611, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1596.4443, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1026.8221, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(732.4316, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(820.6722, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1714.6606, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2058.6609, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1642.9215, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1713.0052, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1759.4592, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1058.2904, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1741.5076, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2178.3918, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1527.4595, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1509.3118, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(492.8512, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1913.6003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1627.5002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1317.7712, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(988.4621, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1379.8719, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(753.9638, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1892.9036, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(975.6093, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1570.5864, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2558.9536, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1721.3116, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2492.3301, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1446.0288, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1271.6362, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1596.2460, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.06060606060606061, 'recall': 0.045454545454545456, 'f1': 0.05194805194805195, 'number': 308}, 'P': {'precision': 0.24643584521384929, 'recall': 0.290167865707434, 'f1': 0.2665198237885462, 'number': 417}, 'overall_precision': 0.18698060941828254, 'overall_recall': 0.18620689655172415, 'overall_f1': 0.18659295093296474, 'overall_accuracy': 0.5659373505499761}
			------------EPOCH 2---------------
Loss:  tensor(1012.0385, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1115.2751, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1230.8022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1604.4604, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(695.1937, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1930.1732, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1668.5356, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1452.6344, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1213.2446, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2119.8123, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1444.1182, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(926.6416, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(662.5407, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1498.4207, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(917.3728, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1497.1294, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1031.8850, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1746.4509, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2760.9382, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2101.5930, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2073.2678, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2030.3090, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(776.8779, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(972.7889, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1801.2360, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2318.9297, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(895.9820, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1192.1609, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(836.4338, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(596.9333, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(660.0752, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1286.7792, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1686.2839, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1359.2675, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1386.1987, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1374.1345, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(848.5118, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1509.7874, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1684.0182, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1230.8662, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1233.9666, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(364.5103, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1619.5889, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1331.0708, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1076.5507, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(800.4949, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1128.2921, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(572.0226, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1550.0227, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(799.4205, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1373.4824, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2095.7974, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1444.4889, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2058.8950, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1284.7063, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1192.9180, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1510.0593, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3194444444444444, 'recall': 0.14935064935064934, 'f1': 0.20353982300884954, 'number': 308}, 'P': {'precision': 0.3803786574870912, 'recall': 0.5299760191846523, 'f1': 0.44288577154308617, 'number': 417}, 'overall_precision': 0.3682758620689655, 'overall_recall': 0.3682758620689655, 'overall_f1': 0.3682758620689655, 'overall_accuracy': 0.6360593017694883}
			------------EPOCH 3---------------
Loss:  tensor(896.1331, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1001.5461, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1036.3356, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1382.8624, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(621.7349, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1606.3759, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1295.6300, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1172.5125, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(890.4454, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1584.5500, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1192.0242, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(795.3555, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(544.4185, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1099.2389, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(758.5144, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1226.3954, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(869.4674, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1328.5088, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2462.4272, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1617.7327, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1661.4971, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1739.8824, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(628.3776, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(843.4061, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1612.0117, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1677.1624, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(758.3887, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(988.6270, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(665.0935, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(475.0313, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(497.3002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1013.4589, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1473.4219, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1130.5211, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1211.4438, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(977.9810, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(650.7728, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1248.1715, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1277.6448, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1012.7725, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1014.6330, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(276.3585, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1477.7339, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1143.3307, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(877.4955, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(631.2173, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(863.8242, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(423.4157, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1287.6359, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(604.7933, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1146.2495, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1585.6155, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1248.6512, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1647.3699, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(972.2817, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(876.7563, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(968.4603, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.25688073394495414, 'recall': 0.18181818181818182, 'f1': 0.2129277566539924, 'number': 308}, 'P': {'precision': 0.39889705882352944, 'recall': 0.5203836930455635, 'f1': 0.45161290322580644, 'number': 417}, 'overall_precision': 0.35826771653543305, 'overall_recall': 0.37655172413793103, 'overall_f1': 0.367182246133154, 'overall_accuracy': 0.6694763271162123}
			------------EPOCH 4---------------
Loss:  tensor(625.7730, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(828.7692, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(773.7973, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1181.4496, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(564.0037, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1295.0214, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1068.7234, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1122.2542, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(719.4971, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1280.2844, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(976.5486, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(663.7532, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(466.8994, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(853.1566, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(590.9891, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1027.2537, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(827.1217, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1061.5096, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1976.9082, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1302.7947, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1281.8040, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1523.8804, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(440.1676, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(821.5079, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1324.6151, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1362.9448, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(573.3995, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(719.8846, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(482.0927, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(375.2223, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(376.5493, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(785.1573, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1298.2667, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1126.1646, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1196.1770, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1044.7153, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(433.0675, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1035.2478, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1000.5216, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(901.6718, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(840.4401, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(188.9478, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1232.7914, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(866.7382, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(694.4268, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(507.2921, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(641.0090, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(324.0014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1120.5034, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(469.1097, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(998.5038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1298.5334, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1150.7388, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1400.4009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(703.0469, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(639.5012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(611.8849, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3979848866498741, 'recall': 0.512987012987013, 'f1': 0.44822695035460997, 'number': 308}, 'P': {'precision': 0.5642201834862385, 'recall': 0.5899280575539568, 'f1': 0.5767878077373974, 'number': 417}, 'overall_precision': 0.4849939975990396, 'overall_recall': 0.5572413793103448, 'overall_f1': 0.5186136071887034, 'overall_accuracy': 0.7160449545671927}
			------------EPOCH 5---------------
Loss:  tensor(576.0705, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(644.0953, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(649.9510, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(968.8005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(429.2219, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(982.6486, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(797.3154, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(905.9670, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(625.9168, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1131.6790, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(870.8431, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(564.8961, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(328.7843, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(665.8923, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(489.4310, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(923.8049, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(554.9883, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(914.8944, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2082.4565, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1178.9342, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1064.3447, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1090.7671, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(358.7635, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(670.9250, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1150.5137, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1425.0945, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(509.6570, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(610.4939, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(498.8655, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(321.6480, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(405.7667, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(671.7475, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1316.2317, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(819.0974, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(884.0294, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(567.8904, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(395.4852, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(993.3524, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(825.1028, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(824.4818, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(688.9474, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(169.9796, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1062.1866, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(695.6699, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(552.2698, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(423.4519, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(511.8914, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(316.0088, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(987.9967, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(430.9203, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(865.5297, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(720.1351, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(997.4105, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1129.1094, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(529.9514, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(489.5348, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(432.3727, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4095744680851064, 'recall': 0.5, 'f1': 0.45029239766081874, 'number': 308}, 'P': {'precision': 0.6099290780141844, 'recall': 0.6187050359712231, 'f1': 0.6142857142857143, 'number': 417}, 'overall_precision': 0.5156445556946183, 'overall_recall': 0.5682758620689655, 'overall_f1': 0.5406824146981628, 'overall_accuracy': 0.7134744141559063}
			------------EPOCH 6---------------
Loss:  tensor(448.0007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(513.9247, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(568.7788, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(846.6383, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(355.3166, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(757.5431, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(613.8843, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(809.3405, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(495.3569, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(806.6875, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(672.5178, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(440.4044, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(312.9059, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(476.3854, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(385.8587, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(651.2448, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(399.1370, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(508.5136, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1208.5188, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(672.7655, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(809.1942, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(784.5322, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(360.2415, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(532.9753, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(793.5613, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(864.1880, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(283.2503, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(493.3233, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(206.4927, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(184.6507, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(203.5086, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(447.7330, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(782.7203, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(625.2589, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(623.3524, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(363.6254, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(186.8559, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(622.5459, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(535.2831, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(613.5492, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(526.7146, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(117.7506, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(745.0189, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(479.0578, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(341.4326, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(284.2650, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(284.1375, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(222.7439, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(739.4467, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(235.9110, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(481.8334, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(366.8926, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(691.7778, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(696.0173, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(267.9514, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(254.1134, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(225.7220, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3944723618090452, 'recall': 0.5097402597402597, 'f1': 0.4447592067988668, 'number': 308}, 'P': {'precision': 0.6336336336336337, 'recall': 0.5059952038369304, 'f1': 0.5626666666666666, 'number': 417}, 'overall_precision': 0.5034199726402189, 'overall_recall': 0.5075862068965518, 'overall_f1': 0.5054945054945055, 'overall_accuracy': 0.6896819703491153}
			------------EPOCH 7---------------
Loss:  tensor(282.1860, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(441.5403, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(363.6841, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(631.0463, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(273.1594, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(372.8157, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(341.3864, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(605.5076, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(343.4379, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(513.2178, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(393.9826, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(277.4416, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(192.2743, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(318.8516, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(229.1040, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(445.6173, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(240.1730, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(263.3132, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(899.5654, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(351.0683, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(560.2417, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(452.1102, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(184.5974, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(278.7576, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(514.6429, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(572.2314, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(146.4084, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(278.8275, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(113.4245, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(116.0622, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(155.8567, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(292.2428, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(470.0775, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(401.2197, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(400.2855, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(243.0209, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(83.0773, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(293.1738, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(311.3162, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(448.7263, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(440.2576, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.8335, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(698.3042, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(453.3337, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(223.4939, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(225.4691, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(204.3040, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(118.7326, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(491.6006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(141.1423, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(273.6613, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(183.3275, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(381.4773, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(339.4294, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(150.5345, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(150.4527, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(144.1690, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.314785373608903, 'recall': 0.6428571428571429, 'f1': 0.4226254002134472, 'number': 308}, 'P': {'precision': 0.76, 'recall': 0.2278177458033573, 'f1': 0.3505535055350553, 'number': 417}, 'overall_precision': 0.3885941644562334, 'overall_recall': 0.40413793103448276, 'overall_f1': 0.39621365787694385, 'overall_accuracy': 0.5942132950741272}
			------------EPOCH 8---------------
Loss:  tensor(270.5998, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(318.2997, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(304.2882, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(441.1949, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(167.4308, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(201.4592, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(279.3684, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(476.8376, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(243.3063, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(392.7425, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(267.7832, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(165.0959, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(129.5286, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(216.0955, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(143.5553, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(248.1013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(239.6281, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(155.3152, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(605.3090, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(249.1716, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(396.9306, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(248.4194, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(112.9647, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(179.0028, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(206.0081, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(464.2087, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(75.1936, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(179.4754, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.4500, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(92.3024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(124.5643, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(169.3518, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(360.6210, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(272.1100, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(276.0514, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(204.1162, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(76.5908, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(160.9480, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(159.8375, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(323.8211, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(280.3426, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.7898, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(420.3822, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(260.1315, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(124.2771, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(125.7503, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(113.7654, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(74.2119, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(510.2083, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(130.4466, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(303.3452, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(160.4792, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(544.4147, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(491.7711, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(192.5510, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(253.3704, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(87.1076, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3326959847036329, 'recall': 0.564935064935065, 'f1': 0.41877256317689526, 'number': 308}, 'P': {'precision': 0.7662337662337663, 'recall': 0.2829736211031175, 'f1': 0.4133099824868651, 'number': 417}, 'overall_precision': 0.43131462333825704, 'overall_recall': 0.4027586206896552, 'overall_f1': 0.41654778887303856, 'overall_accuracy': 0.6065877570540411}
			------------EPOCH 9---------------
Loss:  tensor(203.0026, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(271.8261, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(156.4106, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(301.2432, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(192.8415, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(270.9290, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(291.7123, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(423.7678, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(502.0090, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(840.4020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(766.0022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(443.2886, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(133.9374, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(212.0292, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(219.8956, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(253.2845, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(155.3649, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(108.4182, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(381.5329, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(133.4019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(389.2656, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(407.4626, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(148.9204, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(192.6600, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(444.5697, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(495.2542, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(281.0169, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(380.6803, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(131.9391, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(113.4189, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(155.4828, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(456.3624, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(556.7370, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(572.2927, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(571.8680, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(154.1681, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.4051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(137.2756, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(123.4086, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(314.1580, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(262.2660, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.5610, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(310.4394, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(233.2306, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(249.6888, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(153.6305, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(193.5857, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(250.5069, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(613.8103, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(154.5622, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(534.2565, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(262.6550, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(493.8849, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(413.5834, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(141.6169, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(107.1307, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(103.9660, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.47277936962750716, 'recall': 0.5357142857142857, 'f1': 0.502283105022831, 'number': 308}, 'P': {'precision': 0.5763052208835341, 'recall': 0.6882494004796164, 'f1': 0.6273224043715847, 'number': 417}, 'overall_precision': 0.5336481700118064, 'overall_recall': 0.623448275862069, 'overall_f1': 0.5750636132315522, 'overall_accuracy': 0.7375657580105213}
			------------EPOCH 10---------------
Loss:  tensor(310.4974, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(318.4328, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(354.0774, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(662.6823, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(130.4306, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(448.7379, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(350.8508, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(525.2810, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(213.3028, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(470.4457, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(316.2190, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(151.1939, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(76.1253, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(123.3341, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(76.0745, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(120.9156, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(206.3091, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(210.4936, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(478.1579, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(204.3720, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(523.2457, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(439.0901, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(170.6657, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(237.7497, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(268.3428, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(355.2800, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(122.4036, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(248.4152, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.4908, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(100.5939, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(109.1481, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(204.0627, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(234.0162, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(224.0798, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(199.7791, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(147.3048, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(69.8018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(113.2503, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(118.6435, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(233.1705, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(206.4500, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.7428, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(220.9263, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(164.2137, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(97.9056, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(100.9984, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(92.4655, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(74.2809, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(235.3780, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(90.8650, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(116.1227, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(88.0673, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(175.0039, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(239.3635, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(93.5709, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(142.6543, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(111.3930, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5747126436781609, 'recall': 0.3246753246753247, 'f1': 0.4149377593360996, 'number': 308}, 'P': {'precision': 0.547077922077922, 'recall': 0.8081534772182254, 'f1': 0.6524685382381412, 'number': 417}, 'overall_precision': 0.5531645569620253, 'overall_recall': 0.6027586206896551, 'overall_f1': 0.5768976897689768, 'overall_accuracy': 0.7269846963175514}
			------------EPOCH 11---------------
Loss:  tensor(300.2687, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(283.1132, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(291.0678, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(523.9322, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(75.9666, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(225.4027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(140.5352, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(240.7893, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(178.6528, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(337.0259, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(182.4875, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(84.5711, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.3766, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(74.8710, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.5843, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(84.2180, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(155.4402, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(95.5934, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(282.5660, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(103.2150, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(298.9386, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(272.0558, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(61.5664, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(154.9181, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(207.1056, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(337.9707, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(65.5000, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(178.8342, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.8006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(100.6341, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(144.3781, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(145.6737, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(243.5120, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(285.1146, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(270.1322, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(145.5802, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.7384, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(118.7803, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(182.8725, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(242.3730, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(127.1602, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.2055, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(182.0099, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(126.2615, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(93.9345, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(62.0681, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.4795, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.2377, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(168.1119, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.9758, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(74.0465, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.6348, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(101.4237, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(102.0451, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.0155, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(95.4959, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(75.1993, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5470085470085471, 'recall': 0.2077922077922078, 'f1': 0.30117647058823527, 'number': 308}, 'P': {'precision': 0.5120845921450151, 'recall': 0.8129496402877698, 'f1': 0.6283595922150139, 'number': 417}, 'overall_precision': 0.5173299101412067, 'overall_recall': 0.5558620689655173, 'overall_f1': 0.535904255319149, 'overall_accuracy': 0.7070181731229077}
			------------EPOCH 12---------------
Loss:  tensor(264.1830, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(247.8248, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(233.8847, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(363.1790, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(101.2492, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(204.4728, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(198.0915, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(321.0110, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(228.2805, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(464.6071, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(237.1525, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(174.0056, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.2638, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(157.6155, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.6751, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(144.3719, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(99.5900, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.2999, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(260.6005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(69.3327, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(210.7933, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(141.2077, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.5285, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(82.1272, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(82.9551, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(194.3806, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.2851, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(73.5977, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.6442, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(77.5611, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(134.8727, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(68.8254, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(236.9571, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(240.9808, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(245.5879, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(153.5353, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(126.0856, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(300.7004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(366.6702, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(405.7781, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(262.3950, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(127.3369, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(289.8267, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(326.3202, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(167.6969, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(83.6010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(62.7788, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(88.9631, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(179.6997, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.5884, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(108.0318, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.0331, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(74.6611, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(83.3673, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.1324, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(78.0283, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.8749, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5130434782608696, 'recall': 0.19155844155844157, 'f1': 0.2789598108747045, 'number': 308}, 'P': {'precision': 0.5099667774086378, 'recall': 0.7362110311750599, 'f1': 0.6025515210991168, 'number': 417}, 'overall_precision': 0.5104602510460251, 'overall_recall': 0.5048275862068966, 'overall_f1': 0.507628294036061, 'overall_accuracy': 0.7072572931611669}
			------------EPOCH 13---------------
Loss:  tensor(156.9945, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(145.4429, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(155.9902, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(173.9874, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(78.7627, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(109.6508, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(90.7891, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(113.0761, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(181.5780, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(440.2411, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(304.6378, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(139.4837, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(102.5010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(514.8424, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(383.3765, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(512.3562, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(136.0480, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(367.8536, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1449.9539, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(632.1133, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(254.9115, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(196.9250, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.1216, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(102.6896, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(66.8026, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(195.9888, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.9612, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(65.6609, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.1901, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.6704, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(64.9661, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(68.2042, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(176.4490, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(179.6540, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(180.1315, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(140.8750, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(130.3247, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(231.2819, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(196.6231, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(324.6770, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(297.2084, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(66.8520, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(327.2490, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(217.4786, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(876.4686, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(618.3299, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(770.3746, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(458.3245, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(891.5095, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(138.5015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1317.9890, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(568.3506, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(934.6454, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(491.8439, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(167.8859, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(72.4625, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(91.1285, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.47685185185185186, 'recall': 0.3344155844155844, 'f1': 0.3931297709923664, 'number': 308}, 'P': {'precision': 0.5191489361702127, 'recall': 0.5851318944844125, 'f1': 0.5501691093573845, 'number': 417}, 'overall_precision': 0.5058309037900874, 'overall_recall': 0.4786206896551724, 'overall_f1': 0.4918497519489724, 'overall_accuracy': 0.6958393113342898}
			------------EPOCH 14---------------
Loss:  tensor(198.1336, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(470.3004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(159.1828, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(348.7178, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(126.4269, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(258.9984, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(120.6859, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(370.5670, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(193.3419, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(358.2103, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(275.9392, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(277.8985, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(234.1811, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(402.9733, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(225.9019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(395.7098, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(246.7753, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(206.6414, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(972.5606, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(551.0713, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(745.2006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(592.9440, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(234.5757, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(789.0695, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(511.7910, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(544.5597, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(227.2018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(313.9246, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(68.2879, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(82.1028, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(69.5016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(273.6400, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(318.2899, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(280.0731, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(237.8557, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(114.5075, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(82.5688, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(188.8722, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(128.6483, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(247.0806, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(193.2027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.8409, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(336.5953, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(182.5345, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(168.9156, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(112.8526, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(93.5242, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(62.5851, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(289.2409, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(85.1954, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(195.5279, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(156.0360, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(206.3911, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(207.3435, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(120.5088, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(86.5739, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(134.7236, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.35472370766488415, 'recall': 0.6461038961038961, 'f1': 0.4579976985040276, 'number': 308}, 'P': {'precision': 0.5333333333333333, 'recall': 0.21103117505995203, 'f1': 0.3024054982817869, 'number': 417}, 'overall_precision': 0.3953168044077135, 'overall_recall': 0.39586206896551723, 'overall_f1': 0.39558924879393514, 'overall_accuracy': 0.6113701578192252}
			------------EPOCH 15---------------
Loss:  tensor(338.2983, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(270.9764, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(234.6786, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(361.4691, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(184.4434, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(348.7210, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(256.3048, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(339.5585, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(383.7632, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(905.6216, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(485.8838, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(178.4594, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(151.7153, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(281.2686, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(217.0295, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(291.8723, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(175.5159, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(194.2993, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(390.4981, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(182.0276, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(323.5129, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(139.3710, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(71.7121, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(137.7475, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(72.2076, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(171.8814, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.4216, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(80.7888, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.8646, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.0738, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.9935, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(73.4611, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(244.7149, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(222.8557, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(204.4194, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(95.2950, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.0340, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(333.5755, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(208.8907, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(260.7236, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(249.7823, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.6260, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(733.6883, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(340.5857, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(251.7970, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(186.5794, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(266.5139, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(68.0870, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(331.5241, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(182.4207, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(207.1456, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(353.0729, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(254.8647, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(232.1075, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(115.0631, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(73.1919, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.7102, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5206611570247934, 'recall': 0.6136363636363636, 'f1': 0.5633383010432192, 'number': 308}, 'P': {'precision': 0.5927601809954751, 'recall': 0.6282973621103117, 'f1': 0.6100116414435389, 'number': 417}, 'overall_precision': 0.5602484472049689, 'overall_recall': 0.6220689655172413, 'overall_f1': 0.5895424836601306, 'overall_accuracy': 0.7552606408417025}
			------------EPOCH 16---------------
Loss:  tensor(145.3779, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(114.4926, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(102.4766, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(136.6923, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.3767, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(73.8007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(83.9215, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(85.3649, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(103.0693, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(231.5373, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(97.6970, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.0777, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(65.4830, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(81.4111, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.2228, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.4828, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(110.2281, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(98.2964, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(252.3240, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(135.2545, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(401.4535, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(483.3398, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(245.1225, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(144.5812, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(151.2047, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(201.9465, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.0816, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(155.1163, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.8238, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(72.1596, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.9842, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.4216, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(107.2772, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(85.3658, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(117.6318, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(81.8816, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.5225, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.2921, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.8376, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(129.2154, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(109.6389, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.8334, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(93.2462, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(86.0743, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(93.0798, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(69.3513, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.7272, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.0851, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(171.8715, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(81.6351, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(100.7397, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(76.1582, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(261.5269, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(161.9740, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(67.4189, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(64.1308, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(80.9857, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.40451745379876797, 'recall': 0.6396103896103896, 'f1': 0.4955974842767295, 'number': 308}, 'P': {'precision': 0.6631299734748011, 'recall': 0.5995203836930456, 'f1': 0.6297229219143577, 'number': 417}, 'overall_precision': 0.5173611111111112, 'overall_recall': 0.616551724137931, 'overall_f1': 0.5626179987413468, 'overall_accuracy': 0.7164036346245816}
			------------EPOCH 17---------------
Loss:  tensor(299.7081, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(252.1852, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(141.0333, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(173.9106, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.8615, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(145.0671, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(134.1207, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(160.0507, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(106.1003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(271.0925, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(183.8351, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(64.0708, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.1549, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.7089, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.8542, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(99.7478, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.0031, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.5368, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(101.6748, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.0812, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(173.4307, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(103.3069, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.2952, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(70.7246, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(90.1253, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(141.5287, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.4794, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(83.3818, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.0554, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.3341, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.0297, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(87.7127, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(293.2066, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(189.9768, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(152.2833, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(198.7145, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.7326, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(96.6918, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(83.6042, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(244.5636, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(101.9046, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.3727, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(176.4562, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(137.6380, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(101.6333, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.4861, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(64.0806, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.6151, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(78.8734, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.4339, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.5507, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.4449, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.9522, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(68.0490, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.9051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.7942, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.0638, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4797979797979798, 'recall': 0.6168831168831169, 'f1': 0.5397727272727273, 'number': 308}, 'P': {'precision': 0.6789838337182448, 'recall': 0.7050359712230215, 'f1': 0.6917647058823528, 'number': 417}, 'overall_precision': 0.5838359469240049, 'overall_recall': 0.6675862068965517, 'overall_f1': 0.6229086229086229, 'overall_accuracy': 0.7589670014347202}
			------------EPOCH 18---------------
Loss:  tensor(90.9042, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(77.3760, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.5188, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.0302, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.3255, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.5709, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(61.6059, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.5245, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(70.4772, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(156.9451, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.1625, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.5680, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.3818, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.5240, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.7688, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(87.2369, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(97.8660, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(100.5417, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(169.5790, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(86.8251, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(206.1794, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(272.8215, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.3501, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(246.9452, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.1819, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(116.0853, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.8138, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.0603, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.6819, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.1352, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.7146, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.3299, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.5383, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.5305, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(59.4844, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.4614, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.5024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.0501, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.2589, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(94.4865, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.1446, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.0923, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.1729, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.7329, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.7493, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.0019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.0796, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.4131, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(61.4536, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.2300, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.8389, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.3154, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(87.6728, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.0114, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.9371, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.3017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.5484, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5015873015873016, 'recall': 0.512987012987013, 'f1': 0.507223113964687, 'number': 308}, 'P': {'precision': 0.6437346437346437, 'recall': 0.6282973621103117, 'f1': 0.6359223300970873, 'number': 417}, 'overall_precision': 0.5817174515235457, 'overall_recall': 0.5793103448275863, 'overall_f1': 0.5805114029025571, 'overall_accuracy': 0.7786944045911047}
			------------EPOCH 19---------------
Loss:  tensor(84.2039, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.6895, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.8060, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.7062, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.9809, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.9581, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.3445, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.5989, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.5538, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(111.3351, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.1465, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.9954, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.4562, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.2477, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.8425, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.6070, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.0904, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.8506, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(80.5253, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.2583, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(145.9449, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(74.4822, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.2757, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.4832, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.8100, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(110.0399, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.9738, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.8139, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.3390, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.9024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.9387, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.7620, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.8107, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.1847, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.7317, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.9810, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.6614, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.1245, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.2522, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(79.4664, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.1375, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.6436, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.4818, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.4414, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.4002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.9551, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.4643, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.9626, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.2510, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.7049, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.6551, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.2532, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.3187, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.3580, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.1716, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.1513, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.1513, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5138888888888888, 'recall': 0.6006493506493507, 'f1': 0.5538922155688623, 'number': 308}, 'P': {'precision': 0.6945812807881774, 'recall': 0.6762589928057554, 'f1': 0.6852976913730257, 'number': 417}, 'overall_precision': 0.6096605744125326, 'overall_recall': 0.6441379310344828, 'overall_f1': 0.6264252179745137, 'overall_accuracy': 0.7827594452415112}
			------------EPOCH 20---------------
Loss:  tensor(53.4003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.9419, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.6789, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.7924, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.6828, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.5904, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.0694, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.4976, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.1867, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(84.0213, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.4874, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.9467, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.5493, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.9686, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.9985, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.8867, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.6276, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.5174, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.0879, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.2339, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(124.5986, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.5908, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.9504, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.8058, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.4676, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(91.9719, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.5216, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.3343, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.1890, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.6735, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.1319, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.1116, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.3853, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.1278, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.5120, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.9983, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.3925, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.8134, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.0979, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(67.4907, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.3838, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.7656, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.4696, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.4621, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.7052, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.6596, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.7032, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.4475, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.5789, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.8918, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.5736, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.9788, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.5906, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.3949, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.5001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.0437, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.6598, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5165745856353591, 'recall': 0.6071428571428571, 'f1': 0.5582089552238806, 'number': 308}, 'P': {'precision': 0.6928746928746928, 'recall': 0.6762589928057554, 'f1': 0.6844660194174758, 'number': 417}, 'overall_precision': 0.6098829648894668, 'overall_recall': 0.646896551724138, 'overall_f1': 0.6278447121820616, 'overall_accuracy': 0.7788737446197992}
			------------EPOCH 21---------------
Loss:  tensor(35.3876, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.6848, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.8512, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.4314, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.4801, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.5048, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.2524, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.4109, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.8635, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(75.5472, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.0093, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.0440, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.2988, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.5181, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.6971, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.5255, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.7632, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.5037, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.3640, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.5119, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(112.1973, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.2345, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.7219, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.0436, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.3908, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(85.8823, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.1863, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.0172, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.0059, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.1878, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.0129, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.5197, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.9820, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.3004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.7700, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.5954, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.6730, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.4029, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.5614, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.6569, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.0372, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.7453, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.1412, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.7362, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.5452, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.9128, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.0219, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.1000, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.0390, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.1406, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.8664, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.4816, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.7058, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.7388, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.5791, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.9994, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.7427, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5478260869565217, 'recall': 0.6136363636363636, 'f1': 0.5788667687595711, 'number': 308}, 'P': {'precision': 0.7012345679012346, 'recall': 0.6810551558752997, 'f1': 0.6909975669099757, 'number': 417}, 'overall_precision': 0.6306666666666667, 'overall_recall': 0.6524137931034483, 'overall_f1': 0.6413559322033899, 'overall_accuracy': 0.7826996652319465}
			------------EPOCH 22---------------
Loss:  tensor(29.6713, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.8740, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.3277, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.9219, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.0227, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.6580, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.4830, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.9823, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.9706, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(69.0661, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.9779, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.3917, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.5156, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.8502, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.1346, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.2412, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.5312, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.0251, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.3735, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.0161, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(100.0278, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.2294, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.5336, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.7920, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.7229, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(80.4431, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.8538, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.9974, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.9658, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.7107, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.6502, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.2603, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.7999, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.5990, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.8374, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.0225, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.0895, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.2820, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.7036, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.5513, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.9537, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.8199, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.9923, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.5197, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.3696, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.7274, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.3617, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.6151, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.3714, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.5642, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.7164, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.2513, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.7266, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.9484, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.6348, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.4432, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.7725, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5591715976331361, 'recall': 0.6136363636363636, 'f1': 0.5851393188854489, 'number': 308}, 'P': {'precision': 0.6928746928746928, 'recall': 0.6762589928057554, 'f1': 0.6844660194174758, 'number': 417}, 'overall_precision': 0.6322147651006711, 'overall_recall': 0.6496551724137931, 'overall_f1': 0.6408163265306123, 'overall_accuracy': 0.7842539454806313}
			------------EPOCH 23---------------
Loss:  tensor(24.4508, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.5258, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.0634, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.9179, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.9719, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.1315, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.1190, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.3100, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.4108, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.4331, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.5428, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.2317, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.6136, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.6722, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.8623, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.0366, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.2308, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.2140, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.7668, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.6454, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(90.5036, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.3243, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.0856, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.8699, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.0854, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(75.0201, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.7540, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.6160, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.4116, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.8500, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.7642, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.0182, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.9162, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.7015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.4322, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.5899, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.0669, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.7922, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.5780, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.8345, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.4597, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.1018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.8087, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.6995, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.8189, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.5383, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.6885, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.4370, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.5300, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.4229, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.7577, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.2938, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.0760, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.6333, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.1050, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.4417, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.3101, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5625, 'recall': 0.6136363636363636, 'f1': 0.5869565217391304, 'number': 308}, 'P': {'precision': 0.6887254901960784, 'recall': 0.6738609112709832, 'f1': 0.6812121212121213, 'number': 417}, 'overall_precision': 0.6317204301075269, 'overall_recall': 0.6482758620689655, 'overall_f1': 0.6398910823689584, 'overall_accuracy': 0.7845528455284553}
			------------EPOCH 24---------------
Loss:  tensor(21.5952, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.3221, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.9000, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.6052, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.9717, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.7945, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.3512, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.2503, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.2699, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(59.7487, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.8176, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.3467, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.1929, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.7566, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.8833, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.2419, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.3973, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.9774, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.8597, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.8674, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(82.3789, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.1332, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.8131, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.9734, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.6508, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(70.2917, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.8954, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.6138, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.7068, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.9696, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.2711, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.7731, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.8928, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.1715, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.7245, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.9962, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.4100, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.9970, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.1188, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.4892, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.2655, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.5485, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.2190, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.3056, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.6319, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.3318, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.2812, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.2441, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.1728, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.3869, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.6662, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.4681, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.8307, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.5883, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.0295, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.9810, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.2985, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5608308605341247, 'recall': 0.6136363636363636, 'f1': 0.586046511627907, 'number': 308}, 'P': {'precision': 0.6919315403422983, 'recall': 0.6786570743405276, 'f1': 0.6852300242130751, 'number': 417}, 'overall_precision': 0.6327077747989276, 'overall_recall': 0.6510344827586206, 'overall_f1': 0.6417403127124404, 'overall_accuracy': 0.7849115255858441}
			------------EPOCH 25---------------
Loss:  tensor(19.0149, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.2702, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.0846, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.7333, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.1038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.6545, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.8411, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.6041, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.0655, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.7466, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.3496, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.6312, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.0870, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.9045, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.9157, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.5200, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.6583, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.9264, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.3819, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.5531, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(75.4907, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.2844, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.7821, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.4597, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.5019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(66.1047, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.1116, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.9943, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.3075, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.2325, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.1109, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.7388, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.3250, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.9702, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.2949, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.6070, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.6699, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.3143, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.8092, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.8824, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.3227, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.1242, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.9504, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.2010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.7913, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.9075, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.1485, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.4016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.1513, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.8999, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.0525, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.7264, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.7752, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.5795, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.2821, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.7922, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.5265, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5634218289085545, 'recall': 0.6201298701298701, 'f1': 0.5904173106646059, 'number': 308}, 'P': {'precision': 0.687960687960688, 'recall': 0.6714628297362111, 'f1': 0.6796116504854369, 'number': 417}, 'overall_precision': 0.631367292225201, 'overall_recall': 0.6496551724137931, 'overall_f1': 0.6403806934058465, 'overall_accuracy': 0.7849115255858441}
			------------EPOCH 26---------------
Loss:  tensor(16.7228, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.5150, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.7232, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.2327, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.3794, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.7840, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.5061, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.2497, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.3352, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.1872, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.3445, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.9968, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.1734, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.2067, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.0573, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.9243, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.0074, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.0236, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.0852, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.6205, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(69.2058, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.4372, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.9291, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.2881, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.5640, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(62.1141, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.4270, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.7387, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.3355, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.7307, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.2871, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.8907, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.0568, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.9949, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.1027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.5132, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.0063, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.9390, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.7275, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.7145, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.5043, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.7673, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.8866, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.3207, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.3353, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.0518, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.2068, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.6770, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.3127, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.9150, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.4012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.1201, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.8064, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.5216, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.6854, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.7630, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.8894, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5571847507331378, 'recall': 0.6168831168831169, 'f1': 0.5855161787365176, 'number': 308}, 'P': {'precision': 0.6763990267639902, 'recall': 0.6666666666666666, 'f1': 0.6714975845410628, 'number': 417}, 'overall_precision': 0.6223404255319149, 'overall_recall': 0.6455172413793103, 'overall_f1': 0.6337169939065674, 'overall_accuracy': 0.7843735054997609}
			------------EPOCH 27---------------
Loss:  tensor(14.7591, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.8812, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.6308, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.9614, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.7546, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.0829, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.4029, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.1707, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.9716, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.8215, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.2455, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.4492, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.4297, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.6401, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.2964, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.4688, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.4787, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.3079, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.7772, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.9406, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.4750, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.0296, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.2316, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.4215, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.7486, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.1687, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.8066, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.7969, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.6716, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.4516, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.6907, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.1795, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.0191, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.1845, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.0999, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.6264, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.3834, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.7827, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.8794, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.8506, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.7447, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.4389, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.9124, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.6014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.2253, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.8737, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.4426, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.0686, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.5956, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.2982, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.7392, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.6008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.9637, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.2411, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.1876, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.8470, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.3275, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5558823529411765, 'recall': 0.6136363636363636, 'f1': 0.5833333333333334, 'number': 308}, 'P': {'precision': 0.6747572815533981, 'recall': 0.6666666666666666, 'f1': 0.6706875753920386, 'number': 417}, 'overall_precision': 0.6210106382978723, 'overall_recall': 0.6441379310344828, 'overall_f1': 0.6323628977657414, 'overall_accuracy': 0.7826398852223816}
			------------EPOCH 28---------------
Loss:  tensor(13.4662, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.3365, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.7289, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.8739, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.1959, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.4450, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.4114, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.3036, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.8431, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.6962, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.3781, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.9539, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.8142, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.0950, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.6205, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.0911, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.0408, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.7121, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.8192, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.3711, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.2453, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.8837, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.6448, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.7795, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.9480, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.3225, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.2547, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.0502, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.1755, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.3317, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.2316, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.6267, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.1242, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.5538, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.2436, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.9038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.8549, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.6558, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.1473, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.2369, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.0996, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.1558, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.9992, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.0031, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.3898, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.8478, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.7971, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.5885, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.9483, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.8055, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.2493, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.1554, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.1868, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.7095, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.7678, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.9786, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.8195, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5513196480938416, 'recall': 0.6103896103896104, 'f1': 0.5793528505392912, 'number': 308}, 'P': {'precision': 0.6739659367396593, 'recall': 0.6642685851318945, 'f1': 0.6690821256038646, 'number': 417}, 'overall_precision': 0.6183510638297872, 'overall_recall': 0.6413793103448275, 'overall_f1': 0.6296547054840894, 'overall_accuracy': 0.78162362505978}
			------------EPOCH 29---------------
Loss:  tensor(12.6351, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.8839, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.9736, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.9318, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.6789, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.8775, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.5245, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.6145, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.8855, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.7502, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.8079, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.5087, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.3237, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.6155, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.9891, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.7622, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.6850, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.2004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.0924, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.9148, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.5300, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.9393, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.1417, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.3236, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.3004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.5452, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.7266, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.4362, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.8187, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.2989, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.8910, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.1139, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.3560, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.9841, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.4562, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.2182, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.2965, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.6357, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.6121, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.2721, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.5328, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.8703, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.1222, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.5053, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.7529, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.8994, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.2539, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.2108, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.3262, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.3854, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.8483, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.7641, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.5343, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.4185, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.3912, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.1423, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.3455, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5529411764705883, 'recall': 0.6103896103896104, 'f1': 0.580246913580247, 'number': 308}, 'P': {'precision': 0.6739659367396593, 'recall': 0.6642685851318945, 'f1': 0.6690821256038646, 'number': 417}, 'overall_precision': 0.6191744340878829, 'overall_recall': 0.6413793103448275, 'overall_f1': 0.6300813008130082, 'overall_accuracy': 0.7828192252510761}
			------------EPOCH 30---------------
Loss:  tensor(12.0188, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.5104, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.3319, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.1197, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.1799, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.3650, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.6833, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.0625, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.0463, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.1138, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.3509, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.0982, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.9388, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.1530, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.4571, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.4843, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.3854, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.7634, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.8222, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.4998, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.1722, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.1436, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.7038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.0047, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.6485, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.0219, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.2384, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.9108, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.5275, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.3379, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.6151, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.6742, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.6482, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.5269, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.7147, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.6519, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.8141, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.7907, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.1784, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.1272, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.0522, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.6130, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.2752, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.0646, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.9487, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.7315, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.2486, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.9014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.7763, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.4794, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.3953, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.9003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.9211, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.0766, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.2986, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.9372, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5510204081632653, 'recall': 0.6136363636363636, 'f1': 0.5806451612903225, 'number': 308}, 'P': {'precision': 0.6658595641646489, 'recall': 0.6594724220623501, 'f1': 0.6626506024096385, 'number': 417}, 'overall_precision': 0.6137566137566137, 'overall_recall': 0.64, 'overall_f1': 0.6266036461850102, 'overall_accuracy': 0.7806671449067432}
	Train size: 50 Test size: 50


		-------------RUN 1-----------
			------------EPOCH 1---------------
Loss:  tensor(1678.2830, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2998.6145, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2496.2446, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3422.5894, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2379.2817, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1258.4573, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(955.9185, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1937.5481, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1283.2883, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2161.4785, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2147.1848, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1163.1743, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2017.9003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3055.0757, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1523.4917, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2020.4768, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1963.0078, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2306.1489, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(981.6331, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1930.4497, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2411.7559, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2117.2830, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1477.1902, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2416.7664, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(771.6376, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1251.2170, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1164.2373, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1669.9325, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2006.4968, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3630.5142, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2556.5117, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1744.4333, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1104.3083, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2227.3528, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2063.3274, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'number': 737}, 'P': {'precision': 0.23672566371681417, 'recall': 0.5393145161290323, 'f1': 0.32902829028290287, 'number': 992}, 'overall_precision': 0.23620309050772628, 'overall_recall': 0.3094274146905726, 'overall_f1': 0.2679018527791688, 'overall_accuracy': 0.49409746518294473}
			------------EPOCH 2---------------
Loss:  tensor(1142.2310, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2259.9104, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1944.1820, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2511.2715, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1840.3065, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(952.3026, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(737.0026, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1404.9087, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1007.7867, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1740.2550, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1627.9844, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(867.0167, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1643.1628, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2409.2207, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1197.8654, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1596.7402, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1664.7988, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2031.0229, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(824.7263, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1553.0454, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1976.5251, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1792.3547, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1227.0488, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2130.2810, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(713.4908, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1074.1656, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(975.0095, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1412.8134, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1664.8186, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3010.3159, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2112.2983, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1390.8452, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(921.8629, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1825.7804, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1847.8533, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.2222222222222222, 'recall': 0.04613297150610583, 'f1': 0.07640449438202246, 'number': 737}, 'P': {'precision': 0.28602058319039453, 'recall': 0.6723790322580645, 'f1': 0.40132370637785797, 'number': 992}, 'overall_precision': 0.28209255533199196, 'overall_recall': 0.40543666859456334, 'overall_f1': 0.3327005220692928, 'overall_accuracy': 0.5486535900503352}
			------------EPOCH 3---------------
Loss:  tensor(983.8420, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1963.1332, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1675.8489, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2247.2417, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1654.3506, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(852.3265, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(653.1294, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1219.2612, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(912.8163, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1621.5930, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1466.5552, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(759.3558, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1373.0514, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1842.4812, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(926.5671, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1369.1621, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1324.6709, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1881.1919, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(682.4028, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1187.4420, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1626.2786, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1477.7461, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1011.6051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1775.3376, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(616.3090, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(892.1719, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(808.2032, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1160.5947, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1404.1766, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2756.7805, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1816.8801, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1196.1865, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(774.4775, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1558.2834, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1717.5084, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.14545454545454545, 'recall': 0.17367706919945725, 'f1': 0.15831787260358687, 'number': 737}, 'P': {'precision': 0.3627906976744186, 'recall': 0.6290322580645161, 'f1': 0.4601769911504425, 'number': 992}, 'overall_precision': 0.28923076923076924, 'overall_recall': 0.4349334875650665, 'overall_f1': 0.3474243474243474, 'overall_accuracy': 0.6153533620748031}
			------------EPOCH 4---------------
Loss:  tensor(843.0383, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1649.8862, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1353.8186, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2047.3057, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1474.4514, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(800.8423, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(552.6067, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1013.7944, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(801.4993, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1374.3735, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1247.1876, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(682.0229, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1257.2214, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1638.9971, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(807.3585, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1262.5193, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1168.6912, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1687.0884, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(588.5630, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1050.9712, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1428.1392, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1257.2981, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(804.7210, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1574.6782, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(522.6976, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(782.2510, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(705.0204, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1018.2449, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1159.1641, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2617.2083, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1570.7778, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(943.3686, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(584.2358, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1211.1742, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1526.5846, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.1768219832735962, 'recall': 0.20081411126187246, 'f1': 0.1880559085133418, 'number': 737}, 'P': {'precision': 0.3806413301662708, 'recall': 0.6461693548387096, 'f1': 0.4790732436472346, 'number': 992}, 'overall_precision': 0.3129710432368108, 'overall_recall': 0.45633314054366686, 'overall_f1': 0.3712941176470588, 'overall_accuracy': 0.6243369523508566}
			------------EPOCH 5---------------
Loss:  tensor(707.9853, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1416.8309, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1081.1310, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1841.2369, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1344.4587, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(750.7582, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(504.4341, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(872.5016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(722.0779, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1195.4825, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1032.0200, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(602.9821, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1111.8247, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1402.5175, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(682.6733, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1098.7953, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1066.8784, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1439.8406, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(573.2946, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(964.6503, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1169.0742, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1094.3141, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(638.6652, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1374.1304, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(435.3723, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(636.1426, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(564.4473, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(774.0372, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(886.4929, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2356.4214, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1177.6267, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(644.1711, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(418.2377, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(871.3802, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1336.6143, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.20825515947467166, 'recall': 0.15061058344640435, 'f1': 0.17480314960629922, 'number': 737}, 'P': {'precision': 0.3286573146292585, 'recall': 0.6612903225806451, 'f1': 0.4390896921017403, 'number': 992}, 'overall_precision': 0.3032819296164492, 'overall_recall': 0.44360902255639095, 'overall_f1': 0.3602630342883983, 'overall_accuracy': 0.5869128501455884}
			------------EPOCH 6---------------
Loss:  tensor(631.1475, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1304.5735, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(920.0262, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1788.2617, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1320.5757, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(730.2148, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(498.6564, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(820.1067, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(746.2016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1269.0251, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(962.5410, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(495.7274, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1008.9598, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1184.0481, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(592.0069, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(988.8473, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(880.0994, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1184.4583, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(458.4247, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(799.6968, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(939.4186, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(953.2744, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(505.8937, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1201.7512, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(481.9388, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(643.4014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(499.5741, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(667.1404, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1201.9155, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3155.6309, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1276.5835, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(650.0499, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(337.6605, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(790.6796, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(980.2295, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.26536312849162014, 'recall': 0.12890094979647218, 'f1': 0.17351598173515984, 'number': 737}, 'P': {'precision': 0.4081532416502947, 'recall': 0.8377016129032258, 'f1': 0.5488771466314398, 'number': 992}, 'overall_precision': 0.3868003341687552, 'overall_recall': 0.5355696934644303, 'overall_f1': 0.44918748484113513, 'overall_accuracy': 0.6104552739092161}
			------------EPOCH 7---------------
Loss:  tensor(587.0099, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1126.9979, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(723.0796, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1534.5391, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1237.0422, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(681.1609, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(461.8792, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(761.7258, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(723.5514, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1450.5051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1088.6523, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(432.5832, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1001.5385, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1214.1110, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(661.6429, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1066.3167, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(797.5724, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1617.3269, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(514.3974, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(726.2222, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1309.4320, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1029.5258, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(689.0528, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1741.2354, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(515.3214, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(601.2822, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(491.0535, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(652.2032, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1124.7759, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2456.8491, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1275.6068, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(606.1937, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(424.1230, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(972.7476, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1526.9407, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.29863813229571984, 'recall': 0.41655359565807326, 'f1': 0.3478753541076487, 'number': 737}, 'P': {'precision': 0.46524064171123, 'recall': 0.7016129032258065, 'f1': 0.5594855305466239, 'number': 992}, 'overall_precision': 0.3973851030110935, 'overall_recall': 0.5801041064198958, 'overall_f1': 0.47166705854690805, 'overall_accuracy': 0.6184908471209625}
			------------EPOCH 8---------------
Loss:  tensor(505.0578, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1277.7468, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1277.1663, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1596.7211, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1075.5370, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(566.7938, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(396.5351, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(630.9874, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(588.4263, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1183.7332, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(975.7960, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(415.1081, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(882.4825, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1113.4679, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(489.7394, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(964.6397, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(834.0529, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1258.7911, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(487.8794, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(725.0559, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1105.6328, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1016.7770, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(572.4874, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1282.3187, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(419.4291, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(543.4707, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(419.9439, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(562.9633, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(789.4549, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2003.5699, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(926.9828, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(476.4436, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(354.3584, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(682.4901, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1002.8170, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.2817391304347826, 'recall': 0.4396200814111262, 'f1': 0.34340222575516693, 'number': 737}, 'P': {'precision': 0.4153720359771055, 'recall': 0.5120967741935484, 'f1': 0.45869074492099327, 'number': 992}, 'overall_precision': 0.3506110408765276, 'overall_recall': 0.48120300751879697, 'overall_f1': 0.40565577766942956, 'overall_accuracy': 0.6428458569397106}
			------------EPOCH 9---------------
Loss:  tensor(574.8890, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1050.1451, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(785.9358, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1386.6161, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1051.6598, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(567.7192, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(379.0554, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(561.9408, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(526.8154, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(932.3032, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(734.8087, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(421.0279, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(736.3385, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(851.5587, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(407.0342, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(733.3947, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(572.0182, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(852.7474, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(305.7339, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(536.4709, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(768.0051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(735.1749, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(297.3228, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(954.0598, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(298.2216, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(375.8563, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(262.9338, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(369.0731, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(487.5531, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1816.5554, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(577.4307, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(291.3327, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(225.9556, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(503.4772, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(834.3757, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.37848101265822787, 'recall': 0.4056987788331072, 'f1': 0.39161755075311067, 'number': 737}, 'P': {'precision': 0.4888287068381855, 'recall': 0.7278225806451613, 'f1': 0.5848521668691777, 'number': 992}, 'overall_precision': 0.45037494486104984, 'overall_recall': 0.5905147484094853, 'overall_f1': 0.511011011011011, 'overall_accuracy': 0.657810983454845}
			------------EPOCH 10---------------
Loss:  tensor(309.4773, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(639.8613, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(454.5761, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(907.4581, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(720.1913, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(414.7539, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(259.8656, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(379.9416, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(407.9531, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(827.0244, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(571.7719, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(262.9119, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(515.5799, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(631.4171, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(226.4689, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(540.6459, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(411.9697, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(584.9642, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(207.3329, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(367.5908, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(529.0630, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(617.4193, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(264.9283, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(718.7375, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(263.0764, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(276.7281, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(188.1688, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(224.7142, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(327.8091, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1148.7959, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(360.2791, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(167.4014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(165.7070, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(318.4993, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(571.0175, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.37926136363636365, 'recall': 0.3622795115332429, 'f1': 0.3705759888965996, 'number': 737}, 'P': {'precision': 0.47514204545454547, 'recall': 0.6743951612903226, 'f1': 0.5575, 'number': 992}, 'overall_precision': 0.4431818181818182, 'overall_recall': 0.5413533834586466, 'overall_f1': 0.4873730799271023, 'overall_accuracy': 0.6803376746495723}
			------------EPOCH 11---------------
Loss:  tensor(233.1151, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(478.1477, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(353.8932, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(567.6290, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(514.1573, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(352.7638, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(182.7817, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(296.0289, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(298.5491, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(661.7208, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(470.1396, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(232.5797, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(328.4653, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(503.1174, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(156.5105, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(446.6741, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(254.0783, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(411.6403, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(140.9512, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(257.5140, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(456.6986, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(486.5919, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(169.7203, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(561.3367, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(197.2509, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(225.1505, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(153.7759, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(135.5154, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(194.7291, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(934.8400, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(232.5011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(103.4252, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(128.2620, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(213.7294, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(470.3161, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3868110236220472, 'recall': 0.5332428765264586, 'f1': 0.44837421563034796, 'number': 737}, 'P': {'precision': 0.5205364626990779, 'recall': 0.626008064516129, 'f1': 0.5684210526315789, 'number': 992}, 'overall_precision': 0.4590312358533273, 'overall_recall': 0.5864661654135338, 'overall_f1': 0.5149822244794311, 'overall_accuracy': 0.6706092138229917}
			------------EPOCH 12---------------
Loss:  tensor(120.4029, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(335.9208, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(237.7032, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(392.0309, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(302.2020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(171.6577, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(120.0262, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(205.9140, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(191.8641, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(426.0056, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(313.4418, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(149.8690, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(210.5680, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(434.8883, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(123.5389, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(413.9272, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(206.1381, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(258.9865, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(163.6516, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(222.4246, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(395.5901, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(474.6180, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(147.7419, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(303.0460, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(111.9024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(144.5265, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(95.0278, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(100.1226, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(118.8751, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(487.4522, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(148.0263, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(65.9904, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(92.3192, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(143.5006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(287.9348, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.33163265306122447, 'recall': 0.7055630936227951, 'f1': 0.4511930585683297, 'number': 737}, 'P': {'precision': 0.547923322683706, 'recall': 0.34576612903225806, 'f1': 0.4239802224969097, 'number': 992}, 'overall_precision': 0.39334548769371014, 'overall_recall': 0.4991324465008676, 'overall_f1': 0.43996941116492483, 'overall_accuracy': 0.6308602126266845}
			------------EPOCH 13---------------
Loss:  tensor(189.0445, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(388.6816, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(250.6164, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(485.3301, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(455.6148, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(238.6854, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(191.1645, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(248.9120, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(190.8714, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(267.7477, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(279.2964, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(143.2488, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(104.5942, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(268.0616, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(64.9035, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(247.6711, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(131.5415, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(205.5844, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(71.7130, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(140.5677, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(454.5258, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(471.0467, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(167.3621, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(425.6785, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(125.9888, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(380.7080, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(206.0082, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(162.8069, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(225.1346, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(983.4904, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(252.3927, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(144.9411, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.6088, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(156.2213, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(324.7928, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3440453686200378, 'recall': 0.49389416553595655, 'f1': 0.4055710306406686, 'number': 737}, 'P': {'precision': 0.5141776937618148, 'recall': 0.5483870967741935, 'f1': 0.5307317073170732, 'number': 992}, 'overall_precision': 0.42911153119092627, 'overall_recall': 0.525159051474841, 'overall_f1': 0.47230169050715215, 'overall_accuracy': 0.685619484007855}
			------------EPOCH 14---------------
Loss:  tensor(69.3868, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(195.4437, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(127.5282, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(149.7681, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(177.0084, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(109.9074, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(81.7364, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(123.7862, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(248.8685, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(335.7123, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(486.9961, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(250.7672, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(355.0745, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(705.5455, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(428.7793, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(648.5283, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(326.8694, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(415.2065, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(127.6370, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(365.0270, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(283.6296, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(432.6332, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(99.2519, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(224.7177, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(83.2259, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(85.3020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(81.0804, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(64.7235, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(90.9517, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(377.0742, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(143.1109, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(76.7373, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(91.1799, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(291.2732, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(346.4392, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4223826714801444, 'recall': 0.1587516960651289, 'f1': 0.23076923076923073, 'number': 737}, 'P': {'precision': 0.45570971184631803, 'recall': 0.8608870967741935, 'f1': 0.5959525471039776, 'number': 992}, 'overall_precision': 0.4514179451417945, 'overall_recall': 0.5615962984384038, 'overall_f1': 0.5005154639175258, 'overall_accuracy': 0.6414689750129788}
			------------EPOCH 15---------------
Loss:  tensor(386.9013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(716.8333, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(386.8289, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(722.0537, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(548.7111, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(213.5780, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(174.5453, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(245.2559, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(158.2588, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(298.0856, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(267.3004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(77.4616, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(79.5719, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(255.0934, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.8134, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(185.3701, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(71.7422, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(99.0372, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.7933, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(111.0964, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(198.7297, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(248.9200, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(88.3222, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(216.4213, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(127.8277, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(170.4031, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(93.0730, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(66.9319, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(113.4789, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(458.7890, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(141.3904, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.3450, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.2400, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(112.3759, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(231.0656, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.35372714486638535, 'recall': 0.6824966078697422, 'f1': 0.4659564613246873, 'number': 737}, 'P': {'precision': 0.5278174037089871, 'recall': 0.37298387096774194, 'f1': 0.43709391612522147, 'number': 992}, 'overall_precision': 0.41121055110692417, 'overall_recall': 0.5049161364950838, 'overall_f1': 0.4532710280373832, 'overall_accuracy': 0.65677267905108}
			------------EPOCH 16---------------
Loss:  tensor(83.8033, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(193.9078, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(159.7206, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(205.5919, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(207.5209, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(104.1349, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(92.3798, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(122.9658, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(79.0180, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(126.6195, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(96.0025, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(92.7039, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(74.8631, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(196.5591, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.9800, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(192.1662, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.4189, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(74.4936, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.7244, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(105.3259, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(177.2580, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(169.7699, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(76.1801, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(147.2045, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(61.8734, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(73.6133, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(65.1597, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.7194, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.3954, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(151.0878, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(72.9274, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.6262, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.0156, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.3859, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(76.9555, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4037735849056604, 'recall': 0.4355495251017639, 'f1': 0.41906005221932113, 'number': 737}, 'P': {'precision': 0.5213414634146342, 'recall': 0.6895161290322581, 'f1': 0.59375, 'number': 992}, 'overall_precision': 0.47698149027052683, 'overall_recall': 0.5812608444187392, 'overall_f1': 0.523983315954119, 'overall_accuracy': 0.6809245423560482}
			------------EPOCH 17---------------
Loss:  tensor(37.6922, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(86.7098, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(62.0891, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(77.1035, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(114.5444, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.4039, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.2212, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(75.5074, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(81.3329, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(110.4386, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(81.9519, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.2685, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.5003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(179.2614, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.6981, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(157.5286, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.8810, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(67.7625, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.5528, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(59.2350, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(97.9987, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(156.5143, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.4932, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(62.3240, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.1625, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.4163, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.0298, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.4485, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.3455, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(109.6409, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.8195, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.7858, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.5666, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.4026, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.7026, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.39754601226993863, 'recall': 0.4396200814111262, 'f1': 0.4175257731958763, 'number': 737}, 'P': {'precision': 0.5149884704073789, 'recall': 0.6754032258064516, 'f1': 0.5843872655909289, 'number': 992}, 'overall_precision': 0.46975425330812853, 'overall_recall': 0.5748987854251012, 'overall_f1': 0.5170351105331599, 'overall_accuracy': 0.6786447870347381}
			------------EPOCH 18---------------
Loss:  tensor(28.5087, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(66.6604, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.5227, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(59.8965, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.0750, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.1741, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.8293, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.6803, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.1330, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(67.9072, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.3758, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.6489, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.7246, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(137.2977, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.0046, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(95.2304, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.0752, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.4115, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.1769, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.0786, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(72.0244, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(103.3253, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.2221, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.6248, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.1069, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.5408, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.6128, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.5067, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.3438, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(74.0785, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.7230, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.6635, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.7431, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.3265, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.1406, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3816872427983539, 'recall': 0.5033921302578019, 'f1': 0.43417203042715036, 'number': 737}, 'P': {'precision': 0.5357142857142857, 'recall': 0.6350806451612904, 'f1': 0.5811808118081181, 'number': 992}, 'overall_precision': 0.4660148975791434, 'overall_recall': 0.5789473684210527, 'overall_f1': 0.5163786432808873, 'overall_accuracy': 0.6777870573098888}
			------------EPOCH 19---------------
Loss:  tensor(23.9419, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.3250, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.4720, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.6703, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.9866, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.0120, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.9807, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.8912, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.6191, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.4576, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.3352, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.1547, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.6918, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(117.3208, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.1763, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(81.0484, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.6274, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.7855, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.6756, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.6156, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.8553, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(80.3634, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.3095, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.1629, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.3281, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.4722, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.6852, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.0633, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.4169, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.7325, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.8083, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.1724, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.3762, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.9422, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.1986, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.39955357142857145, 'recall': 0.48575305291723203, 'f1': 0.43845682792406615, 'number': 737}, 'P': {'precision': 0.5434782608695652, 'recall': 0.655241935483871, 'f1': 0.5941499085923218, 'number': 992}, 'overall_precision': 0.4818355640535373, 'overall_recall': 0.582995951417004, 'overall_f1': 0.5276105731483904, 'overall_accuracy': 0.6844457485949033}
			------------EPOCH 20---------------
Loss:  tensor(21.0949, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.3285, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.4014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.0055, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.3237, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.4710, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.9870, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.5316, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.4821, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.8011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.0686, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.8451, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.7712, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(102.1022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.5316, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(67.6196, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.4585, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.4203, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.8254, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.7469, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.6547, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(67.3845, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.0062, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.5803, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.8148, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.6074, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.4902, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.8255, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.0911, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.0112, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.1711, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.9447, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.9455, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.1980, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.9388, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4068965517241379, 'recall': 0.48032564450474896, 'f1': 0.4405724953329185, 'number': 737}, 'P': {'precision': 0.543046357615894, 'recall': 0.6612903225806451, 'f1': 0.5963636363636364, 'number': 992}, 'overall_precision': 0.48604427333974976, 'overall_recall': 0.5841526894158473, 'overall_f1': 0.5306015235093249, 'overall_accuracy': 0.68458117960409}
			------------EPOCH 21---------------
Loss:  tensor(19.2898, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.9528, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.8111, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.5667, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.2069, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.4248, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.4491, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.0881, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.2656, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.9180, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.1047, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.0648, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.2671, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(87.2544, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.0500, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(59.0753, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.9665, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.8678, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.1599, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.3235, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.3655, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.3733, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.2391, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.2856, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.5746, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.8708, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.1695, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.7032, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.8096, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.2637, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.0265, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.6072, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.9738, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.1703, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.4796, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4096662830840046, 'recall': 0.4830393487109905, 'f1': 0.4433374844333749, 'number': 737}, 'P': {'precision': 0.5429524603836531, 'recall': 0.65625, 'f1': 0.5942492012779552, 'number': 992}, 'overall_precision': 0.48694390715667313, 'overall_recall': 0.5824175824175825, 'overall_f1': 0.5304187516460365, 'overall_accuracy': 0.6845134640994966}
			------------EPOCH 22---------------
Loss:  tensor(17.4800, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.0554, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.4942, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.6391, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.2740, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.6565, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.5284, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.0294, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.1116, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.9076, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.0502, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.6024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.1348, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(74.6135, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.4103, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.0734, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.6051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.7601, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.0006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.5378, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.1205, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.6352, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.6991, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.3043, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.9310, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.7094, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.5165, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.1852, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.2000, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.9249, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.4620, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.9108, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.8427, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.6524, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.8559, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.40397350993377484, 'recall': 0.4966078697421981, 'f1': 0.44552647595861233, 'number': 737}, 'P': {'precision': 0.5453774385072095, 'recall': 0.6481854838709677, 'f1': 0.5923537540304007, 'number': 992}, 'overall_precision': 0.4839328537170264, 'overall_recall': 0.5835743204164257, 'overall_f1': 0.5291033036182485, 'overall_accuracy': 0.6868157912556712}
			------------EPOCH 23---------------
Loss:  tensor(15.2397, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.6626, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.2468, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.1140, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.4509, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.0643, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.4232, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.3039, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.2498, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.6993, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.0424, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.2269, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.2921, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(67.1284, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.8679, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.5978, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.9972, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.9712, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.8665, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.1238, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.2794, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.9055, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.3331, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.3000, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.5461, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.7425, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.2217, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.1027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.8190, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.7547, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.1704, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.4589, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.3766, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.3058, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.4669, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.40350877192982454, 'recall': 0.4993215739484396, 'f1': 0.44633110976349305, 'number': 737}, 'P': {'precision': 0.5448334756618275, 'recall': 0.6431451612903226, 'f1': 0.5899214054553861, 'number': 992}, 'overall_precision': 0.4829572731637062, 'overall_recall': 0.5818392134181608, 'overall_f1': 0.527806925498426, 'overall_accuracy': 0.6869963659345868}
			------------EPOCH 24---------------
Loss:  tensor(13.3471, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.6903, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.3655, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.6315, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.0575, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.6134, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.1711, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.0185, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.7980, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.5021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.8094, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.0874, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.0636, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.9922, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.1879, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.4817, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.6110, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.2556, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.5107, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.9471, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.0179, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.0644, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.9398, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.5981, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.2720, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.7906, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.1031, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.0508, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.1087, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.4672, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.7229, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.3065, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.2672, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.1839, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.1450, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.40264026402640263, 'recall': 0.4966078697421981, 'f1': 0.4447144592952612, 'number': 737}, 'P': {'precision': 0.5426621160409556, 'recall': 0.6411290322580645, 'f1': 0.5878003696857671, 'number': 992}, 'overall_precision': 0.4814992791926958, 'overall_recall': 0.5795257374204743, 'overall_f1': 0.5259842519685041, 'overall_accuracy': 0.6862740672189243}
			------------EPOCH 25---------------
Loss:  tensor(11.8188, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.3229, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.3053, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.4797, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.5334, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.3519, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.7976, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.2488, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.8781, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.9192, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.5682, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.2125, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.1158, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.9123, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.7915, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.3722, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.2983, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.5435, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.4977, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.9845, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.1616, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.6935, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.8937, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.1002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.3466, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.9101, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.0582, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.0829, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.2049, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.8791, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.0038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.4869, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.4179, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.2033, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.0574, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4024122807017544, 'recall': 0.49796472184531887, 'f1': 0.4451182534869618, 'number': 737}, 'P': {'precision': 0.5402397260273972, 'recall': 0.6360887096774194, 'f1': 0.5842592592592594, 'number': 992}, 'overall_precision': 0.4798076923076923, 'overall_recall': 0.5772122614227877, 'overall_f1': 0.5240220530322919, 'overall_accuracy': 0.6864997855675687}
			------------EPOCH 26---------------
Loss:  tensor(10.5545, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.5213, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.3207, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.3437, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.4699, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.2435, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.6951, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.6535, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.9548, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.5949, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.2233, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.3253, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.1717, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.6326, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.7688, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.9375, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.0309, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.1137, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.8499, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.2700, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.5047, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.7886, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.3287, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.6593, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.4352, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.0728, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.9737, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.1502, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.5505, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.1065, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.1649, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.9381, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.7886, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.3683, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.1019, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4008714596949891, 'recall': 0.4993215739484396, 'f1': 0.4447129909365559, 'number': 737}, 'P': {'precision': 0.5411663807890223, 'recall': 0.6360887096774194, 'f1': 0.5848007414272475, 'number': 992}, 'overall_precision': 0.4793666026871401, 'overall_recall': 0.5777906304222094, 'overall_f1': 0.5239968528717545, 'overall_accuracy': 0.6859129178610929}
			------------EPOCH 27---------------
Loss:  tensor(9.1945, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.7556, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.1834, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.4842, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.5935, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.2810, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.4879, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.3685, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.3134, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.4023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.3400, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.5760, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.4993, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.2372, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.7007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.7358, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.3293, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.0295, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.4367, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.9946, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.0265, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.0819, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.6987, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.5499, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.8540, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.1395, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.6457, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.0448, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.9871, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.8110, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.7529, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.5242, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.3012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.5420, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.2184, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4046306504961411, 'recall': 0.49796472184531887, 'f1': 0.4464720194647202, 'number': 737}, 'P': {'precision': 0.5452991452991452, 'recall': 0.6431451612903226, 'f1': 0.5901942645698427, 'number': 992}, 'overall_precision': 0.4838709677419355, 'overall_recall': 0.5812608444187392, 'overall_f1': 0.5281135049921177, 'overall_accuracy': 0.6880346703383518}
			------------EPOCH 28---------------
Loss:  tensor(7.5702, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.5095, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.1291, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.7913, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.6519, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.2419, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.0451, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.3818, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.7161, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.4465, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.6533, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.0032, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.8588, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.8685, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.7169, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.7690, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.0396, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.1703, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.1900, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.9720, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.6237, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.4559, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.4372, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.3644, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.4101, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.9355, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.1039, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.3776, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.5436, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.6178, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.3718, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.1761, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.8363, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.7003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.4879, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.40465631929046564, 'recall': 0.49525101763907736, 'f1': 0.44539353264185483, 'number': 737}, 'P': {'precision': 0.5440677966101695, 'recall': 0.6471774193548387, 'f1': 0.5911602209944752, 'number': 992}, 'overall_precision': 0.48366954851104704, 'overall_recall': 0.5824175824175825, 'overall_f1': 0.5284702177906062, 'overall_accuracy': 0.6879669548337585}
			------------EPOCH 29---------------
Loss:  tensor(6.7057, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.2129, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.4431, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.1634, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.7582, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.4190, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.7718, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.4908, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.2779, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.4954, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.0922, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.3365, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.2115, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.3091, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.0103, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.4739, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.2804, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.5058, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.9342, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.2535, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.3452, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.0056, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.5570, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.5314, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.2246, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.0303, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.8416, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.1748, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.0383, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.3087, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.2485, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.8576, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.3903, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.8857, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.8725, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4057649667405765, 'recall': 0.4966078697421981, 'f1': 0.4466137888956681, 'number': 737}, 'P': {'precision': 0.5407905803195963, 'recall': 0.6481854838709677, 'f1': 0.5896377808344796, 'number': 992}, 'overall_precision': 0.48254423720707795, 'overall_recall': 0.5835743204164257, 'overall_f1': 0.5282722513089004, 'overall_accuracy': 0.687650949145656}
			------------EPOCH 30---------------
Loss:  tensor(6.1822, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.2893, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.2556, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.5931, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.2984, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.8303, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.0137, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.6975, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.8009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.7372, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.3809, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.7651, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.6449, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.2188, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.5258, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.2330, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.6209, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.8969, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.6048, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.5579, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.7981, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.9285, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.7096, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.5563, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.0375, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.3807, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.8099, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.4365, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.6482, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.8985, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.7701, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.5357, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.9196, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.1445, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.3789, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.40555555555555556, 'recall': 0.49525101763907736, 'f1': 0.44593769089798413, 'number': 737}, 'P': {'precision': 0.534453781512605, 'recall': 0.6411290322580645, 'f1': 0.5829514207149405, 'number': 992}, 'overall_precision': 0.4789473684210526, 'overall_recall': 0.5789473684210527, 'overall_f1': 0.5242210002618487, 'overall_accuracy': 0.6879669548337585}


		-------------RUN 2-----------
			------------EPOCH 1---------------
Loss:  tensor(1583.6890, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3272.3413, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1415.5380, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2700.0979, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2854.7188, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2832.9648, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3024.5083, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2681.7007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2042.0452, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2010.4486, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1377.0652, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2258.7305, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1493.7518, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1759.0620, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1545.8555, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2413.0032, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2376.9988, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1296.4106, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1653.8186, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2036.3857, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1330.9038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1382.1223, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1424.0380, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2120.9297, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(986.5950, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1706.7151, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1396.6396, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2058.5835, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1346.5291, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1924.2500, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(889.6050, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2163.6631, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2345.4873, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1049.0725, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.08771929824561403, 'recall': 0.024330900243309004, 'f1': 0.0380952380952381, 'number': 822}, 'P': {'precision': 0.08627678054429647, 'recall': 0.1565126050420168, 'f1': 0.11123553564762971, 'number': 952}, 'overall_precision': 0.08644501278772379, 'overall_recall': 0.09526493799323563, 'overall_f1': 0.09064092249932959, 'overall_accuracy': 0.5091808222823435}
			------------EPOCH 2---------------
Loss:  tensor(1101.4161, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2289.2715, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(943.4767, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1835.7809, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2173.2244, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2134.7808, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2258.5894, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2008.1489, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1605.2888, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1644.5151, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1008.2727, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1712.5203, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1190.9149, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1428.0151, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1242.9819, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1938.7258, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2043.7887, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1154.5601, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1505.7549, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1894.3802, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1142.6870, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1160.1169, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1145.7311, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1631.4182, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(788.6039, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1405.5510, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1135.1219, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1730.3743, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1085.7430, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1599.9697, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(694.7065, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1943.8862, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2093.7500, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(916.1016, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.0998389694041868, 'recall': 0.07542579075425791, 'f1': 0.08593208593208593, 'number': 822}, 'P': {'precision': 0.25405405405405407, 'recall': 0.3949579831932773, 'f1': 0.3092105263157895, 'number': 952}, 'overall_precision': 0.20847215611613518, 'overall_recall': 0.24689966178128522, 'overall_f1': 0.22606451612903225, 'overall_accuracy': 0.5738900962434027}
			------------EPOCH 3---------------
Loss:  tensor(976.2156, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1844.9247, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(775.1038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1598.6361, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1925.5917, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1915.5051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1970.8641, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1594.7991, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1357.8967, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1371.6188, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(847.4212, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1455.2834, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1030.1538, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1282.8652, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1003.1329, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1586.8600, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1663.5786, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(895.8358, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1123.6696, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1450.7114, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(974.3670, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(970.0953, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(945.9394, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1353.8958, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(703.2836, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1234.4109, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(977.3204, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1434.4292, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(900.4966, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1387.8768, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(596.7460, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1716.1736, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1803.6659, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(827.1209, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.12634408602150538, 'recall': 0.11435523114355231, 'f1': 0.12005108556832694, 'number': 822}, 'P': {'precision': 0.35013812154696133, 'recall': 0.532563025210084, 'f1': 0.4225, 'number': 952}, 'overall_precision': 0.27417883211678834, 'overall_recall': 0.338782412626832, 'overall_f1': 0.30307614725163895, 'overall_accuracy': 0.6000133055395396}
			------------EPOCH 4---------------
Loss:  tensor(895.5782, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1546.0984, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(654.2324, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1409.8884, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1692.9467, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1697.5800, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1719.0465, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1189.5060, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1082.7458, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1048.6328, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(687.3237, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1196.1952, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(923.2544, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1152.3496, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(834.3562, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1349.4683, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1384.0492, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(673.4570, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(781.4646, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1095.0190, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(884.5522, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(826.4652, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(767.7488, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1078.9900, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(606.2424, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1049.1028, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(813.8914, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1024.6826, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(667.1671, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1099.8572, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(461.7619, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1408.9694, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1545.8402, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(754.8586, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.16007194244604317, 'recall': 0.10827250608272507, 'f1': 0.12917271407837444, 'number': 822}, 'P': {'precision': 0.337736932797243, 'recall': 0.6176470588235294, 'f1': 0.4366877088748607, 'number': 952}, 'overall_precision': 0.29473225946887244, 'overall_recall': 0.38162344983089064, 'overall_f1': 0.33259641365757797, 'overall_accuracy': 0.5704084800638666}
			------------EPOCH 5---------------
Loss:  tensor(822.5458, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1392.5886, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(554.8588, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1325.2773, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1516.8311, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1491.7478, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1545.8394, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1012.6179, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(891.8345, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(754.5056, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(536.1536, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(908.5629, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(802.8532, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(987.9630, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(763.8638, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1215.1243, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1215.5330, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(506.6146, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(567.8395, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(857.6752, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(772.5795, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(693.1315, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(643.5083, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1001.1139, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(607.1376, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1010.0758, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(804.2909, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1094.2833, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(657.1411, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1017.0414, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(500.9378, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1317.0520, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1231.0253, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(632.6412, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.19781931464174454, 'recall': 0.15450121654501217, 'f1': 0.17349726775956284, 'number': 822}, 'P': {'precision': 0.3537170263788969, 'recall': 0.6197478991596639, 'f1': 0.450381679389313, 'number': 952}, 'overall_precision': 0.3103896103896104, 'overall_recall': 0.4041713641488162, 'overall_f1': 0.351126346718903, 'overall_accuracy': 0.5825165210449283}
			------------EPOCH 6---------------
Loss:  tensor(724.9265, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1199.4436, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(447.1758, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1057.7499, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1484.2466, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1478.8350, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1554.0283, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1020.0270, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(988.3450, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(683.8749, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(562.9269, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(987.9534, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(877.4680, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1027.1797, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(744.5247, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1035.4792, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1106.2495, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(555.0875, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(522.6843, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(786.4155, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(673.0776, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(630.2905, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(621.0696, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(930.6125, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(530.9725, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(855.4279, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(739.6010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(757.5176, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(645.7175, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(879.5360, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(497.7596, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1259.0439, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1352.3015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(746.8099, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.1713381555153707, 'recall': 0.4610705596107056, 'f1': 0.24983520105471324, 'number': 822}, 'P': {'precision': 0.49872773536895676, 'recall': 0.20588235294117646, 'f1': 0.2914498141263941, 'number': 952}, 'overall_precision': 0.22072936660268713, 'overall_recall': 0.3241262683201804, 'overall_f1': 0.26261703585293444, 'overall_accuracy': 0.5492748480950903}
			------------EPOCH 7---------------
Loss:  tensor(785.5460, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1499.6815, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(569.0236, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1206.4087, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1497.6238, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1390.6456, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1570.5699, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1263.9855, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(816.7325, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(643.4020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(439.6788, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(729.1656, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(592.2330, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(757.1827, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(545.1993, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(820.5291, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(926.7465, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(448.5082, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(415.5577, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(662.0201, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(671.0969, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(668.7059, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(764.1754, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1109.0789, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(505.7361, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(770.4000, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(681.1360, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(781.3988, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(482.2502, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(831.2050, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(414.1518, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(891.6635, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1017.3362, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(532.5464, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.27699530516431925, 'recall': 0.21532846715328466, 'f1': 0.24229979466119095, 'number': 822}, 'P': {'precision': 0.46002460024600245, 'recall': 0.7857142857142857, 'f1': 0.5802948021722266, 'number': 952}, 'overall_precision': 0.4083885209713024, 'overall_recall': 0.5214205186020293, 'overall_f1': 0.45803416687298837, 'overall_accuracy': 0.6240298044085688}
			------------EPOCH 8---------------
Loss:  tensor(510.9918, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(997.1211, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(355.1205, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(736.7332, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1068.3915, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1002.6589, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1032.7610, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(657.1329, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(618.9907, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(383.8041, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(315.2446, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(528.5044, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(442.8547, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(598.7405, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(419.7570, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(645.5580, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(683.3730, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(248.8670, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(273.9063, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(383.6052, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(419.4658, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(399.7406, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(423.4911, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(663.2352, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(369.1763, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(505.8966, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(423.3154, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(415.5111, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(442.6832, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(489.7285, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(314.0016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(679.5427, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(700.1472, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(315.7285, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.34619750283768447, 'recall': 0.3710462287104623, 'f1': 0.35819142689371697, 'number': 822}, 'P': {'precision': 0.5376068376068376, 'recall': 0.6607142857142857, 'f1': 0.592836946277097, 'number': 952}, 'overall_precision': 0.45538761579717213, 'overall_recall': 0.5264937993235626, 'overall_f1': 0.4883660130718954, 'overall_accuracy': 0.6804674679558256}
			------------EPOCH 9---------------
Loss:  tensor(402.8085, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(783.1961, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(306.9539, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(463.2585, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(792.9432, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(655.2881, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(808.6792, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(420.0805, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(523.8672, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(247.3378, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(248.9966, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(477.9134, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(445.2853, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(535.5227, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(333.6844, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(495.4604, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(479.3193, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(205.1971, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(211.3687, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(313.1256, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(266.6561, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(317.2032, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(372.7241, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(544.0389, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(248.2159, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(316.3273, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(256.0691, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(285.6212, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(323.6404, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(422.0520, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(260.0924, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(492.5057, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(502.1797, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(209.3468, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.2833333333333333, 'recall': 0.3309002433090024, 'f1': 0.30527497194163866, 'number': 822}, 'P': {'precision': 0.4956140350877193, 'recall': 0.47478991596638653, 'f1': 0.48497854077253216, 'number': 952}, 'overall_precision': 0.38675213675213677, 'overall_recall': 0.4081172491544532, 'overall_f1': 0.39714755896873283, 'overall_accuracy': 0.6739921053798732}
			------------EPOCH 10---------------
Loss:  tensor(318.7158, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(854.3975, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(286.2826, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(415.6765, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(570.5278, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(519.3623, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(598.1401, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(286.8384, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(283.8243, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(150.8668, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(170.7289, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(268.4314, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(272.1223, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(286.8058, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(229.5899, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(321.5391, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(458.6364, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(177.6009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(198.7203, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(328.0698, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(372.3011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(430.9404, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(507.2385, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(624.8232, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(234.9907, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(305.9779, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(286.5518, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(353.2743, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(258.0168, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(357.2995, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(193.8991, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(502.7582, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(446.2729, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(223.9025, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3103154305200341, 'recall': 0.44282238442822386, 'f1': 0.3649122807017544, 'number': 822}, 'P': {'precision': 0.5704365079365079, 'recall': 0.6039915966386554, 'f1': 0.586734693877551, 'number': 952}, 'overall_precision': 0.4305364511691884, 'overall_recall': 0.5293122886133033, 'overall_f1': 0.4748419721871049, 'overall_accuracy': 0.6786933960172085}
			------------EPOCH 11---------------
Loss:  tensor(159.7965, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(534.1084, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(167.8959, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(188.1284, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(449.3688, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(385.5645, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(406.7619, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(159.7158, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(274.0209, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(271.4075, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(162.5039, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(367.3004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(493.3175, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(468.8485, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(304.7294, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(588.9944, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(354.2245, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(107.2708, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(179.7862, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(147.2155, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(200.2404, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(219.0089, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(207.5389, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(243.6831, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(137.0196, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(139.8746, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(118.7955, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(165.8277, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(235.7144, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(326.0923, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(150.1148, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(431.0329, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(655.9277, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(184.9276, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3333333333333333, 'recall': 0.33698296836982966, 'f1': 0.3351482153660012, 'number': 822}, 'P': {'precision': 0.48664886515353806, 'recall': 0.7657563025210085, 'f1': 0.5951020408163266, 'number': 952}, 'overall_precision': 0.43194504079003865, 'overall_recall': 0.5670800450958287, 'overall_f1': 0.49037289787960026, 'overall_accuracy': 0.6261365148356766}
			------------EPOCH 12---------------
Loss:  tensor(399.9070, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(668.1385, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(148.2221, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(459.8414, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(471.6458, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(440.6482, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(575.3474, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(242.8720, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(190.3839, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(76.2794, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(105.9110, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(164.3587, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(204.6288, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(155.1486, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(144.8744, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(209.7611, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(180.5075, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(71.4640, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(103.8657, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(103.9123, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(188.0606, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(174.0769, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(149.1889, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(195.9538, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(138.9864, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(80.1794, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(184.6235, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(152.4438, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(173.0962, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(181.9704, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(129.7512, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(301.1605, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(357.1940, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(144.7529, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3372502937720329, 'recall': 0.3491484184914842, 'f1': 0.34309623430962344, 'number': 822}, 'P': {'precision': 0.5310283687943262, 'recall': 0.6292016806722689, 'f1': 0.5759615384615384, 'number': 952}, 'overall_precision': 0.4477008590197069, 'overall_recall': 0.49943630214205187, 'overall_f1': 0.47215560884625635, 'overall_accuracy': 0.7033973477624518}
			------------EPOCH 13---------------
Loss:  tensor(89.7344, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(305.8062, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(124.8865, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(80.1349, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(149.3597, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(163.3285, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(201.8253, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(79.9530, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(74.0710, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.1187, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(66.0266, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(93.3257, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(121.3003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(99.1716, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(100.9434, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(157.9124, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(121.6788, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.9010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.2940, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(67.4857, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(128.9032, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(114.9954, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(98.8591, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(122.0207, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(64.6250, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.9385, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(66.2353, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(67.3949, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(98.3290, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(114.6758, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.5230, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(173.4275, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(193.3624, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.3361, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3657587548638132, 'recall': 0.45742092457420924, 'f1': 0.4064864864864865, 'number': 822}, 'P': {'precision': 0.5613577023498695, 'recall': 0.6775210084033614, 'f1': 0.6139933365064255, 'number': 952}, 'overall_precision': 0.468994028479559, 'overall_recall': 0.5755355129650507, 'overall_f1': 0.5168311819792457, 'overall_accuracy': 0.6868319510356145}
			------------EPOCH 14---------------
Loss:  tensor(62.0152, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(215.6188, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(91.0154, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.5640, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(78.3146, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(89.9656, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(109.9576, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.4732, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.2157, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.0972, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.4442, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(73.1924, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(93.3284, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.7108, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(72.9264, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(131.0153, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(83.4138, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.6099, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.6031, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.4169, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(77.8677, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(78.9747, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(71.6704, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(84.3600, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.8597, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.1017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.9336, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.4880, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(66.3285, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(87.7898, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.2191, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(82.4042, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(82.4339, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.2258, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3922413793103448, 'recall': 0.44282238442822386, 'f1': 0.416, 'number': 822}, 'P': {'precision': 0.5581589958158996, 'recall': 0.7006302521008403, 'f1': 0.6213320912901723, 'number': 952}, 'overall_precision': 0.48563353744700893, 'overall_recall': 0.5811724915445321, 'overall_f1': 0.5291249679240441, 'overall_accuracy': 0.6921541668514658}
			------------EPOCH 15---------------
Loss:  tensor(36.6014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(123.4253, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(64.6037, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.1372, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.6360, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.2141, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(91.3176, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.3455, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.7817, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.3743, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.5959, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.6727, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(65.8012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.1497, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.6579, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(127.3213, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(83.0975, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.0691, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.1024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.5447, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(90.6993, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.5156, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.3916, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.3851, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.4842, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.1538, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.4662, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.1574, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.4690, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(65.8933, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.2967, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(62.5535, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.4945, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.9015, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.40558510638297873, 'recall': 0.3710462287104623, 'f1': 0.38754764930114355, 'number': 822}, 'P': {'precision': 0.5271739130434783, 'recall': 0.7132352941176471, 'f1': 0.6062500000000001, 'number': 952}, 'overall_precision': 0.4823529411764706, 'overall_recall': 0.5546786922209695, 'overall_f1': 0.5159937073938123, 'overall_accuracy': 0.6935290726038941}
			------------EPOCH 16---------------
Loss:  tensor(25.1610, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(93.9411, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(59.0335, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.9643, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.9352, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.3675, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(64.4118, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.7685, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.9617, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.3123, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.1673, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.8533, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(130.5380, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.4137, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(64.8209, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(166.2576, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(80.7705, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.2902, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.8903, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.4796, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.0474, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.1901, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.5658, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.0667, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.1848, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.6509, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.8220, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.9216, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.8954, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.6031, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.8534, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.6743, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.7739, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.8058, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4014839241549876, 'recall': 0.5924574209245742, 'f1': 0.47862407862407863, 'number': 822}, 'P': {'precision': 0.5668449197860963, 'recall': 0.5567226890756303, 'f1': 0.5617382087970324, 'number': 952}, 'overall_precision': 0.473463687150838, 'overall_recall': 0.5732807215332582, 'overall_f1': 0.5186129525752167, 'overall_accuracy': 0.6727724309220738}
			------------EPOCH 17---------------
Loss:  tensor(26.2331, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(97.2709, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.2333, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.6385, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.6200, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.7631, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.4054, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.8250, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.0908, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.5990, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.3330, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.3144, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.5771, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.3833, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.2233, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(112.0008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.8838, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.1413, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.6096, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.9847, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.7450, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.7911, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.6558, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.3888, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.9533, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.2082, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.2586, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.6121, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.7138, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.5650, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.2047, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.0232, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.5648, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.3827, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4485842026825633, 'recall': 0.3661800486618005, 'f1': 0.4032150033489618, 'number': 822}, 'P': {'precision': 0.5211062590975255, 'recall': 0.7521008403361344, 'f1': 0.6156491831470335, 'number': 952}, 'overall_precision': 0.49731051344743277, 'overall_recall': 0.5732807215332582, 'overall_f1': 0.5326001571091908, 'overall_accuracy': 0.6871645895241052}
			------------EPOCH 18---------------
Loss:  tensor(19.2472, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(110.0544, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.6174, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.7473, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.7629, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.1974, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(96.6692, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.5226, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.2474, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.2178, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.6173, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.9387, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.5818, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.0278, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.4260, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(117.4816, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.1541, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.2363, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.5916, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.7835, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.5424, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.9090, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.4367, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.1246, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.4695, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.0266, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.2513, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.5290, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.4844, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.3294, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.4881, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.6013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.9214, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.8549, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.44387229660144184, 'recall': 0.524330900243309, 'f1': 0.48075850529838254, 'number': 822}, 'P': {'precision': 0.5478260869565217, 'recall': 0.6617647058823529, 'f1': 0.5994291151284491, 'number': 952}, 'overall_precision': 0.5002357378595003, 'overall_recall': 0.5980834272829764, 'overall_f1': 0.544801026957638, 'overall_accuracy': 0.6901805118197543}
			------------EPOCH 19---------------
Loss:  tensor(17.7923, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.9209, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.2938, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.8072, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.2121, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.5678, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.6992, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.9506, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.9676, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.0698, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.5281, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.9440, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.4323, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.3234, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.6651, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(81.2293, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.1846, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.0638, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.5891, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.2504, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.2782, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.0083, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.0037, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.8240, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.8509, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.4645, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.1511, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.0026, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.4157, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.1861, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.4946, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.8584, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.9514, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.8629, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4547437295528899, 'recall': 0.5072992700729927, 'f1': 0.4795859689476711, 'number': 822}, 'P': {'precision': 0.5431893687707641, 'recall': 0.6869747899159664, 'f1': 0.6066790352504637, 'number': 952}, 'overall_precision': 0.504950495049505, 'overall_recall': 0.6037204058624577, 'overall_f1': 0.5499358151476251, 'overall_accuracy': 0.6959018938217945}
			------------EPOCH 20---------------
Loss:  tensor(15.1584, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.1745, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.4788, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.4790, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.3729, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.9079, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.8152, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.4307, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.1983, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.9474, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.6902, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.2164, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.2043, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.9588, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.9857, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.4322, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.8233, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.1591, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.2617, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.4979, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.3235, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.5214, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.9100, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.7772, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.7241, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.4083, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.0361, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.8216, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.3718, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.8013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.2119, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.5700, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.4873, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.8109, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.46830985915492956, 'recall': 0.4854014598540146, 'f1': 0.47670250896057353, 'number': 822}, 'P': {'precision': 0.5487012987012987, 'recall': 0.7100840336134454, 'f1': 0.619047619047619, 'number': 952}, 'overall_precision': 0.5158349328214972, 'overall_recall': 0.6059751972942503, 'overall_f1': 0.5572835666148264, 'overall_accuracy': 0.6915332416729498}
			------------EPOCH 21---------------
Loss:  tensor(13.3871, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.9017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.8449, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.2461, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.2661, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.8222, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.4902, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.6858, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.5889, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.1430, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.5274, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.5724, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.3171, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.3461, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.1848, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.3193, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.9137, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.4364, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.3312, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.1340, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.7983, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.1698, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.7607, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.1974, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.6422, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.3035, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.0413, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.3776, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.6152, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.4486, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.4059, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.7827, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.2786, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.0491, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.46320346320346323, 'recall': 0.5206812652068127, 'f1': 0.4902634593356243, 'number': 822}, 'P': {'precision': 0.5528109028960818, 'recall': 0.6817226890756303, 'f1': 0.6105362182502352, 'number': 952}, 'overall_precision': 0.5133460438512869, 'overall_recall': 0.6071025930101466, 'overall_f1': 0.5563016528925621, 'overall_accuracy': 0.692065463254535}
			------------EPOCH 22---------------
Loss:  tensor(12.0185, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.8463, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.4217, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.1496, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.1935, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.5239, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.3375, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.9851, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.0051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.8926, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.6639, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.7915, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.1394, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.0172, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.9394, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.8599, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.9790, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.9897, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.4257, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.6152, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.9136, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.6155, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.6228, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.0881, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.9955, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.9396, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.3828, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.3810, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.9429, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.7964, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.7606, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.4393, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.1941, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.8874, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4629237288135593, 'recall': 0.5316301703163017, 'f1': 0.4949037372593431, 'number': 822}, 'P': {'precision': 0.5537261698440208, 'recall': 0.6712184873949579, 'f1': 0.6068376068376069, 'number': 952}, 'overall_precision': 0.5128693994280267, 'overall_recall': 0.6065388951521984, 'overall_f1': 0.5557851239669421, 'overall_accuracy': 0.6936843038985231}
			------------EPOCH 23---------------
Loss:  tensor(10.9744, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.8967, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.0030, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.2485, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.1745, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.3597, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.3825, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.3463, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.8539, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.5599, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.6343, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.7909, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.9791, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.7454, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.6666, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.8253, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.6842, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.5553, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.7061, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.0488, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.0577, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.2563, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.2546, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.9012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.2027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.7054, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.2847, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.3955, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.2697, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.1200, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.9446, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.5745, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.5438, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.9997, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4614572333685322, 'recall': 0.5316301703163017, 'f1': 0.49406444318824194, 'number': 822}, 'P': {'precision': 0.5559440559440559, 'recall': 0.6680672268907563, 'f1': 0.6068702290076335, 'number': 952}, 'overall_precision': 0.5131516021042564, 'overall_recall': 0.604847801578354, 'overall_f1': 0.5552393272962485, 'overall_accuracy': 0.6923537499445602}
			------------EPOCH 24---------------
Loss:  tensor(9.2622, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.6590, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.8441, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.9500, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.0671, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.0744, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.4263, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.2043, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.5539, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(61.3732, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.3627, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.3784, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.5429, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.6592, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.4276, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.1642, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.1550, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.8050, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.7433, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.2351, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.5952, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.6603, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.9753, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.3198, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.3010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.1331, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.6389, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.3937, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.1248, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.3036, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.6518, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.8614, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.0081, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.2628, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.47368421052631576, 'recall': 0.5036496350364964, 'f1': 0.4882075471698113, 'number': 822}, 'P': {'precision': 0.533955857385399, 'recall': 0.6607142857142857, 'f1': 0.5906103286384976, 'number': 952}, 'overall_precision': 0.5082846003898636, 'overall_recall': 0.5879368658399098, 'overall_f1': 0.5452169367485624, 'overall_accuracy': 0.6947487470616933}
			------------EPOCH 25---------------
Loss:  tensor(12.5984, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.9577, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.7536, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.9619, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.5169, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.8673, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.4688, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.5068, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.0987, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.7341, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.5416, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.9060, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.9272, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.7897, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.1906, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(62.0653, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.1781, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.6549, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.4062, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.8924, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.8139, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.0281, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.0351, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.7331, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.4993, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.2573, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.0603, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.1281, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.8305, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.5467, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.3325, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.6616, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.5026, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.1902, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.48423423423423423, 'recall': 0.5231143552311436, 'f1': 0.5029239766081871, 'number': 822}, 'P': {'precision': 0.5426829268292683, 'recall': 0.6544117647058824, 'f1': 0.5933333333333333, 'number': 952}, 'overall_precision': 0.5171905697445972, 'overall_recall': 0.5935738444193912, 'overall_f1': 0.552755905511811, 'overall_accuracy': 0.6963454118064487}
			------------EPOCH 26---------------
Loss:  tensor(10.2607, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.8203, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.7133, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.0839, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.3056, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.5074, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.9016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.2074, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.2673, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.9475, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.5923, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.6632, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.4806, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.3260, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.6956, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.0031, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.6415, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.0361, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.2927, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.3476, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.4422, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.2459, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.5578, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.2631, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.6590, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.8051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.3099, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.1050, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.2482, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.5086, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.4663, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.3529, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.7782, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.9046, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5, 'recall': 0.5340632603406326, 'f1': 0.5164705882352941, 'number': 822}, 'P': {'precision': 0.560458958517211, 'recall': 0.667016806722689, 'f1': 0.6091127098321343, 'number': 952}, 'overall_precision': 0.5340626553953257, 'overall_recall': 0.6054114994363021, 'overall_f1': 0.5675033025099075, 'overall_accuracy': 0.702532487692376}
			------------EPOCH 27---------------
Loss:  tensor(8.7077, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.0826, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.4190, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.0758, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.3547, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.0487, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.8468, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.1207, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.0365, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.2213, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.2893, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.0293, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.5860, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.5583, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.6630, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.6024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.9731, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.8251, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.9713, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.7660, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.7638, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.3202, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.8596, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.5274, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.5127, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.7604, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.9612, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.8007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.9764, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.0634, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.8966, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.7304, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.4698, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.4377, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.48368953880764903, 'recall': 0.5231143552311436, 'f1': 0.5026300409117475, 'number': 822}, 'P': {'precision': 0.5505716798592788, 'recall': 0.657563025210084, 'f1': 0.5993298228817616, 'number': 952}, 'overall_precision': 0.5212240868706811, 'overall_recall': 0.5952649379932357, 'overall_f1': 0.5557894736842105, 'overall_accuracy': 0.7015567481261366}
			------------EPOCH 28---------------
Loss:  tensor(7.5608, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.1060, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.2352, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.5406, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.1743, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.2903, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.8769, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.3168, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.6462, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.6681, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.5778, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.0335, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.0890, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.8593, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.2078, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.6262, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.2860, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.5822, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.4755, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.8971, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.2336, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.9697, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.1134, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.6502, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.2244, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.7065, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.1106, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.1247, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.7050, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.7876, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.1316, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.2422, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.2583, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.7756, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4820065430752454, 'recall': 0.537712895377129, 'f1': 0.508338125359402, 'number': 822}, 'P': {'precision': 0.5548672566371682, 'recall': 0.6586134453781513, 'f1': 0.6023054755043228, 'number': 952}, 'overall_precision': 0.5222276502198339, 'overall_recall': 0.6025930101465614, 'overall_f1': 0.5595393875948704, 'overall_accuracy': 0.6999157315829156}
			------------EPOCH 29---------------
Loss:  tensor(6.3591, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.0050, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.2083, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.2677, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.6354, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.6270, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.9867, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.7485, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.8376, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.8065, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.3206, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.3826, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.7241, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.4513, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.7694, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.2860, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.2579, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.5540, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.8887, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.4046, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.2663, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.3062, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.0142, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.7040, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.7175, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.9833, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.9468, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.9072, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.9971, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.2570, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.3314, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.2819, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.3703, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.2718, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.48035914702581367, 'recall': 0.5206812652068127, 'f1': 0.49970811441914764, 'number': 822}, 'P': {'precision': 0.5466210436270317, 'recall': 0.6712184873949579, 'f1': 0.6025459688826025, 'number': 952}, 'overall_precision': 0.5179611650485437, 'overall_recall': 0.6014656144306652, 'overall_f1': 0.5565988523735002, 'overall_accuracy': 0.6955249035348383}
			------------EPOCH 30---------------
Loss:  tensor(5.3949, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.4033, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.9069, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.2370, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.5725, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.9281, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.9943, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.8327, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.5891, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.8993, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.8040, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.2436, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.9314, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.6334, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.6571, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.1586, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.8308, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.9670, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.9829, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.2563, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.2295, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.1224, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.6518, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.2933, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.6915, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.0908, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.1768, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.0828, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.3652, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.7118, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.7035, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.9522, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.6897, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.6228, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.48236632536973834, 'recall': 0.5158150851581509, 'f1': 0.4985302763080541, 'number': 822}, 'P': {'precision': 0.5478632478632479, 'recall': 0.6733193277310925, 'f1': 0.6041470311027334, 'number': 952}, 'overall_precision': 0.5197657393850659, 'overall_recall': 0.6003382187147689, 'overall_f1': 0.5571540674862673, 'overall_accuracy': 0.6941943495808756}


		-------------RUN 3-----------
			------------EPOCH 1---------------
Loss:  tensor(2743.2651, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4122.7207, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2180.8140, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1367.5555, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1734.9320, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2853.9609, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3189.5007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1862.4788, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1309.8779, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2526.3091, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1246.3894, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1566.5444, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1791.9917, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1275.3860, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2170.9961, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2142.2888, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1109.8597, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2046.1344, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2700.4509, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2913.7773, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3266.0078, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2205.9785, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2194.9395, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1536.7061, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2245.9180, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2153.4395, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1598.1758, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2825.4812, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1347.4642, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2124.6743, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(886.1516, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1191.8984, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1220.4357, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1850.0414, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2273.6643, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1750.8231, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1612.5752, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'number': 635}, 'P': {'precision': 0.02511415525114155, 'recall': 0.026602176541717048, 'f1': 0.02583675866118614, 'number': 827}, 'overall_precision': 0.023182297154899896, 'overall_recall': 0.015047879616963064, 'overall_f1': 0.018249688925756947, 'overall_accuracy': 0.5026216845282246}
			------------EPOCH 2---------------
Loss:  tensor(1613.0616, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2577.9590, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1715.8960, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1047.0356, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1446.7297, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2370.8901, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2579.0107, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1365.2627, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(995.5699, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1888.9385, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(996.8409, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1217.0740, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1441.1111, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1032.5465, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1819.3075, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1643.3223, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(867.6873, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1607.1570, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2142.8347, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2244.5044, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2653.4561, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1735.9535, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1867.3633, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1170.1541, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1927.2930, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1766.5372, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1322.2917, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2300.2451, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1152.2316, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1881.7085, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(763.1131, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1016.6547, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(978.9589, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1742.9250, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1887.8853, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1410.3772, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1387.2300, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.1309823677581864, 'recall': 0.16377952755905512, 'f1': 0.14555633310006996, 'number': 635}, 'P': {'precision': 0.17755856966707767, 'recall': 0.17412333736396615, 'f1': 0.1758241758241758, 'number': 827}, 'overall_precision': 0.1545171339563863, 'overall_recall': 0.16963064295485636, 'overall_f1': 0.16172155200521685, 'overall_accuracy': 0.5973859886947848}
			------------EPOCH 3---------------
Loss:  tensor(1429.3502, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2372.6621, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1249.5610, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(785.3859, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1049.1919, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1835.1863, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2016.9324, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(947.5512, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(797.6096, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1505.4664, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(811.2977, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(967.0535, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1223.9907, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(876.9796, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1527.8639, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1390.0862, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(722.7574, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1385.8372, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1664.1572, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1849.2087, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2337.3655, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1491.1791, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1672.2013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(952.9166, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1524.0591, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1349.4225, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1065.4431, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1625.9077, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(808.2294, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1384.1281, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(657.0604, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(874.3313, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(894.8251, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1375.4521, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1421.3284, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1191.3390, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1217.6329, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.24587912087912087, 'recall': 0.28188976377952757, 'f1': 0.26265590608950845, 'number': 635}, 'P': {'precision': 0.414070351758794, 'recall': 0.49818621523579204, 'f1': 0.4522502744237102, 'number': 827}, 'overall_precision': 0.343006384213581, 'overall_recall': 0.4042407660738714, 'overall_f1': 0.3711145996860283, 'overall_accuracy': 0.6552932450060107}
			------------EPOCH 4---------------
Loss:  tensor(1101.8011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1868.8911, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(955.1721, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(620.9948, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(703.7432, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1506.4492, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1730.4823, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(761.9270, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(651.4291, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1300.7861, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(693.8159, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(802.9106, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1045.8711, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(728.7440, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1260.6274, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1221.0284, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(606.7494, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1187.3672, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1457.3138, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1653.7646, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1914.9070, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1229.5439, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1348.6572, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(799.6978, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1267.1580, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1026.3618, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(913.2010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1359.8715, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(668.2941, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1256.6569, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(546.8766, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(706.2067, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(613.0726, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1114.5986, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1195.2598, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(954.4338, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(978.5284, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.21107266435986158, 'recall': 0.1921259842519685, 'f1': 0.20115416323165702, 'number': 635}, 'P': {'precision': 0.34311224489795916, 'recall': 0.3252720677146312, 'f1': 0.3339540657976412, 'number': 827}, 'overall_precision': 0.28707782672540383, 'overall_recall': 0.26744186046511625, 'overall_f1': 0.2769121813031162, 'overall_accuracy': 0.5937795738803489}
			------------EPOCH 5---------------
Loss:  tensor(902.0040, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1772.5748, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(634.2046, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(539.1292, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(441.8283, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1197.8879, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1463.9144, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(605.7546, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(458.5830, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1043.8458, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(537.7981, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(675.1431, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(918.3074, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(613.7851, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1111.6193, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1025.9807, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(465.7659, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1069.0475, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1451.3884, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1340.0077, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1552.5657, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(965.5870, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1165.5706, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(652.3100, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(880.3087, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(831.7742, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(703.7508, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1089.0500, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(575.0558, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(905.7148, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(487.6487, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(519.0115, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(683.9373, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(937.4852, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(957.2592, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(825.4077, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(763.4545, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.30935251798561153, 'recall': 0.47401574803149604, 'f1': 0.37437810945273636, 'number': 635}, 'P': {'precision': 0.5282167042889391, 'recall': 0.56590084643289, 'f1': 0.5464098073555166, 'number': 827}, 'overall_precision': 0.41366325981710594, 'overall_recall': 0.5259917920656635, 'overall_f1': 0.4631135200240891, 'overall_accuracy': 0.6819193288487607}
			------------EPOCH 6---------------
Loss:  tensor(718.6494, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1319.4541, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(439.0818, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(344.0322, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(317.0336, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(987.5803, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1219.2341, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(438.0688, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(318.1194, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(812.6213, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(399.5815, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(563.6799, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(675.6837, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(409.4288, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(914.7500, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(899.1263, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(353.9763, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(766.6064, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1037.6558, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1101.8611, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1168.9344, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(730.5521, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(911.1141, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(549.6486, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(679.8286, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(603.2259, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(601.9422, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(939.3457, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(459.2570, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(836.6671, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(324.2869, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(436.6265, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(417.8954, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(847.9841, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(873.1501, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(692.4102, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(578.5809, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3340782122905028, 'recall': 0.47086614173228347, 'f1': 0.3908496732026143, 'number': 635}, 'P': {'precision': 0.5483405483405484, 'recall': 0.45949214026602175, 'f1': 0.5, 'number': 827}, 'overall_precision': 0.4275818639798489, 'overall_recall': 0.46443228454172364, 'overall_f1': 0.44524590163934424, 'overall_accuracy': 0.6707675780750442}
			------------EPOCH 7---------------
Loss:  tensor(478.7600, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(878.1780, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(290.0511, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(267.0850, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(224.7631, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(651.7821, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(925.4854, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(321.7951, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(212.3215, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(745.1809, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(304.3236, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(534.8030, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(503.7235, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(335.4095, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(683.5272, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(682.5443, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(299.2279, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(575.3427, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(756.3632, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(813.4995, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(746.5773, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(542.9215, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(801.9235, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(458.3104, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(455.9843, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(431.1669, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(440.6130, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(758.5654, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(294.8393, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(486.2728, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(243.1304, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(317.4928, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(293.8177, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(603.7933, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(658.2233, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(488.5597, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(521.3624, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.32996389891696754, 'recall': 0.7196850393700788, 'f1': 0.45247524752475254, 'number': 635}, 'P': {'precision': 0.5340599455040872, 'recall': 0.2370012091898428, 'f1': 0.3283082077051926, 'number': 827}, 'overall_precision': 0.3727168949771689, 'overall_recall': 0.4466484268125855, 'overall_f1': 0.4063472308649657, 'overall_accuracy': 0.5892779497148118}
			------------EPOCH 8---------------
Loss:  tensor(544.1207, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1100.2681, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(253.6267, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(257.4900, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(167.6334, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(451.3099, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(757.0969, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(303.8813, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(172.2708, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(509.6596, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(230.7982, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(365.6970, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(487.3065, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(282.1375, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(713.4537, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(598.6028, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(286.4918, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(519.0963, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(698.3672, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(830.6580, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1072.1494, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(645.9960, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(793.3066, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(508.4166, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(349.6180, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(421.0253, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(428.5430, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(679.5571, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(263.5469, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(463.5287, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(250.4713, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(390.3799, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(316.7374, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(518.1622, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(615.7289, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(504.2889, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(449.4510, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.28413284132841327, 'recall': 0.7275590551181103, 'f1': 0.40866873065015474, 'number': 635}, 'P': {'precision': 0.609271523178808, 'recall': 0.22249093107617895, 'f1': 0.3259521700620018, 'number': 827}, 'overall_precision': 0.3350622406639004, 'overall_recall': 0.4418604651162791, 'overall_f1': 0.3811209439528023, 'overall_accuracy': 0.5284548686599995}
			------------EPOCH 9---------------
Loss:  tensor(554.4301, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(874.6346, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(289.6319, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(226.5599, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(237.1046, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(671.8903, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(855.3959, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(377.0454, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(452.2271, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1085.9939, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(519.2819, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(627.9014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(704.7222, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(482.6839, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(777.6070, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(722.7258, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(323.2585, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(721.7900, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(893.6399, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(977.3513, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1071.6849, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(435.2704, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(647.5665, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(584.7249, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(374.4941, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(452.6354, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(373.9584, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(605.9196, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(311.1531, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(494.3102, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(277.5091, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(517.7682, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(731.0526, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(859.0723, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1128.9502, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1295.9580, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1121.3192, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3465909090909091, 'recall': 0.09606299212598425, 'f1': 0.15043156596794083, 'number': 635}, 'P': {'precision': 0.4263565891472868, 'recall': 0.8645707376058042, 'f1': 0.5710862619808307, 'number': 827}, 'overall_precision': 0.4187803561791689, 'overall_recall': 0.53077975376197, 'overall_f1': 0.4681749622926094, 'overall_accuracy': 0.5951607540220477}
			------------EPOCH 10---------------
Loss:  tensor(772.7598, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1818.8152, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(904.3180, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(642.5916, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(473.2000, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1706.6521, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2105.5972, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1115.6128, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(495.6612, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1132.4961, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(544.7461, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(776.6266, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(638.5774, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(390.6261, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(928.0774, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(788.1958, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(308.3714, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(639.8625, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(929.4825, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(748.0798, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(970.8979, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(608.3811, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(703.0186, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(549.2734, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(938.5500, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(732.8547, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(544.7630, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1226.8677, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(610.1151, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(993.1007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(412.5312, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(563.5089, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(537.4318, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1289.0433, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1171.8044, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(666.6786, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(856.8853, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.30474840538625086, 'recall': 0.6771653543307087, 'f1': 0.42033235581622674, 'number': 635}, 'P': {'precision': 0.42058823529411765, 'recall': 0.17291414752116083, 'f1': 0.24507283633247648, 'number': 827}, 'overall_precision': 0.32724157624214734, 'overall_recall': 0.39192886456908343, 'overall_f1': 0.35667600373482733, 'overall_accuracy': 0.5894058367649692}
			------------EPOCH 11---------------
Loss:  tensor(591.3792, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1074.8138, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(261.7841, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(257.0144, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(234.1687, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(546.4047, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(777.9159, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(300.2018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(261.8530, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(640.8591, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(297.4132, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(436.6082, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(493.2446, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(340.7175, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(592.1786, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(566.7462, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(297.1522, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(623.2291, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(853.0231, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(641.3434, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(741.2371, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(484.5755, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(639.8674, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(446.5350, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(525.1799, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(412.0762, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(397.6578, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(647.4897, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(247.1242, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(359.5262, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(212.7406, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(268.7412, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(275.9935, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(450.5014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(558.2823, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(407.4023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(457.3313, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.47754137115839246, 'recall': 0.31811023622047246, 'f1': 0.38185255198487716, 'number': 635}, 'P': {'precision': 0.4883720930232558, 'recall': 0.7110036275695284, 'f1': 0.5790251107828656, 'number': 827}, 'overall_precision': 0.4855562384757222, 'overall_recall': 0.5403556771545828, 'overall_f1': 0.5114923923599871, 'overall_accuracy': 0.7130726142670794}
			------------EPOCH 12---------------
Loss:  tensor(321.5428, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(720.8491, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(235.0805, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(254.4127, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(162.3968, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(612.1565, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(695.4310, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(243.2572, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(173.1836, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(463.3978, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(223.6960, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(313.4665, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(316.4943, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(221.2977, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(502.6857, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(460.9771, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(161.0273, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(311.7166, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(395.7625, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(405.8586, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(418.2295, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(237.1634, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(402.7195, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(261.3198, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(264.7379, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(228.2913, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(257.9768, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(445.8676, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(173.7597, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(243.3251, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(196.1873, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(193.4409, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(182.2357, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(369.7519, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(384.7242, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(303.6276, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(311.5435, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4126637554585153, 'recall': 0.5952755905511811, 'f1': 0.4874274661508704, 'number': 635}, 'P': {'precision': 0.6015424164524421, 'recall': 0.56590084643289, 'f1': 0.5831775700934578, 'number': 827}, 'overall_precision': 0.49940968122786306, 'overall_recall': 0.5786593707250342, 'overall_f1': 0.5361216730038023, 'overall_accuracy': 0.7137120495178658}
			------------EPOCH 13---------------
Loss:  tensor(200.4178, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(318.9127, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(124.5803, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(165.3002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(79.4883, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(247.9249, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(297.2173, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(128.0760, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(111.0699, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(271.2226, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(122.9398, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(181.3811, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(223.5885, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(196.6629, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(276.9359, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(257.0425, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(133.6702, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(198.2466, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(322.8973, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(263.4970, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(259.8455, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(143.8986, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(312.8249, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(211.3971, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(141.9101, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(142.4839, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(185.2088, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(307.5043, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(91.6884, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(129.7816, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(119.9997, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(103.4884, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(95.2263, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(194.3630, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(293.7063, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(166.4310, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(222.8208, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4845679012345679, 'recall': 0.494488188976378, 'f1': 0.48947778643803586, 'number': 635}, 'P': {'precision': 0.5896130346232179, 'recall': 0.7001209189842805, 'f1': 0.6401326699834163, 'number': 827}, 'overall_precision': 0.5478527607361963, 'overall_recall': 0.6108071135430917, 'overall_f1': 0.5776196636481241, 'overall_accuracy': 0.736066705885362}
			------------EPOCH 14---------------
Loss:  tensor(155.2960, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(259.3380, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(109.8381, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(135.3031, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(64.0448, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(220.1128, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(226.9169, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(59.9451, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(61.0485, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(224.0530, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(67.2756, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(116.2482, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(160.5815, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(152.8100, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(254.9207, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(166.8937, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(92.6409, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(113.2771, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(225.9685, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(174.8297, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(167.0654, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(110.0837, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(191.6743, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(181.8297, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(92.7550, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(92.3058, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(167.4019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(260.1040, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(75.5627, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(103.8773, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(92.8461, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(75.5985, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.6411, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(165.3266, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(228.7308, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(114.4378, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(126.6504, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.43411764705882355, 'recall': 0.5811023622047244, 'f1': 0.49696969696969695, 'number': 635}, 'P': {'precision': 0.5895249695493301, 'recall': 0.585247883917775, 'f1': 0.587378640776699, 'number': 827}, 'overall_precision': 0.5104727707959306, 'overall_recall': 0.5834473324213406, 'overall_f1': 0.5445260134056814, 'overall_accuracy': 0.7154768908100366}
			------------EPOCH 15---------------
Loss:  tensor(122.2392, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(145.7001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(83.1505, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(79.3249, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.8586, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(116.5850, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(136.2237, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.9992, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.2721, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(145.5994, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.1822, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(79.4486, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(115.3485, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(96.9177, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(170.4363, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(101.9844, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(64.6393, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(78.2038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(171.1459, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(103.7755, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(112.8640, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(71.2884, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(122.0344, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(112.7537, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(62.2635, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.6311, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(109.8322, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(200.2340, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.0728, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(71.7792, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(74.0740, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.3425, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.6543, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(97.5327, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(153.0993, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(77.6110, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(89.6992, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.47790055248618785, 'recall': 0.5448818897637795, 'f1': 0.5091979396615158, 'number': 635}, 'P': {'precision': 0.588495575221239, 'recall': 0.6432889963724304, 'f1': 0.6146735990756788, 'number': 827}, 'overall_precision': 0.5393120393120393, 'overall_recall': 0.600547195622435, 'overall_f1': 0.5682847896440129, 'overall_accuracy': 0.7269867253241936}
			------------EPOCH 16---------------
Loss:  tensor(109.4819, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(108.0090, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(73.9093, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.9771, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.4719, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(87.0326, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(98.7040, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.4790, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.0975, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(104.4704, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.2337, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(61.8299, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(82.7329, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(64.7201, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(124.4559, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(70.9117, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.4509, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.8197, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(128.0366, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(79.1312, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(74.3986, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.6142, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(78.3998, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(83.7442, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.5969, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.8038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(89.5770, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(166.6573, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.8273, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.4282, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(65.1151, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.9371, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.4311, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(75.1568, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(107.5369, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.9020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.1983, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.46465968586387435, 'recall': 0.5590551181102362, 'f1': 0.5075053609721228, 'number': 635}, 'P': {'precision': 0.5891829689298044, 'recall': 0.619105199516324, 'f1': 0.6037735849056604, 'number': 827}, 'overall_precision': 0.5309246785058175, 'overall_recall': 0.5930232558139535, 'overall_f1': 0.5602584814216478, 'overall_accuracy': 0.7224595237486252}
			------------EPOCH 17---------------
Loss:  tensor(97.1130, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(85.5184, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.7034, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.9478, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.3016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(67.6749, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(82.0060, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.1496, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.4049, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(64.5992, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.6172, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.4363, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.7286, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.3835, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(92.7446, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.1245, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.2625, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.7587, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(105.8131, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(66.5718, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.1776, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.2073, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.8556, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.3898, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.5204, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.4348, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(72.7998, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(145.1241, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.7139, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.7388, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.4513, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.9681, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.1719, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.1877, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(97.2533, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.5573, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.5693, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4746666666666667, 'recall': 0.5606299212598426, 'f1': 0.5140794223826715, 'number': 635}, 'P': {'precision': 0.5936435868331441, 'recall': 0.6324062877871826, 'f1': 0.6124121779859485, 'number': 827}, 'overall_precision': 0.5389331698344574, 'overall_recall': 0.6012311901504788, 'overall_f1': 0.5683802133850631, 'overall_accuracy': 0.7244034069110162}
			------------EPOCH 18---------------
Loss:  tensor(88.9846, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(73.6833, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.5505, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.4299, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.6920, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.6657, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(72.9460, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.2765, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.5736, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.2631, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.4807, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.3609, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.4966, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.0634, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(71.2569, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.4398, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.6285, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.2939, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(92.2363, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.8147, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.1929, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.0467, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.0702, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.3793, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.2095, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.3577, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(62.1243, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(130.5722, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.7100, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.7515, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.0366, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.0510, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.0236, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.8153, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(75.0385, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.7519, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.4079, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.480946123521682, 'recall': 0.5763779527559055, 'f1': 0.5243553008595988, 'number': 635}, 'P': {'precision': 0.5960648148148148, 'recall': 0.6227327690447401, 'f1': 0.6091070372560615, 'number': 827}, 'overall_precision': 0.5421538461538462, 'overall_recall': 0.6025991792065664, 'overall_f1': 0.5707806932296728, 'overall_accuracy': 0.7223827915185308}
			------------EPOCH 19---------------
Loss:  tensor(83.9391, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(66.3515, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.6861, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.8382, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.8752, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.2621, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(67.3885, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.9184, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.6107, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.8735, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.8514, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.3126, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.9030, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.1850, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.9352, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.1839, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.2309, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.3099, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(80.4906, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.7706, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.9956, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.0082, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.7145, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.4002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.3345, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.1777, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.5687, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(117.5324, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.4981, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.4428, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.3946, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.3261, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.3753, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.8843, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.9038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.6118, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.0305, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4885906040268456, 'recall': 0.573228346456693, 'f1': 0.527536231884058, 'number': 635}, 'P': {'precision': 0.5931818181818181, 'recall': 0.6311970979443773, 'f1': 0.6115992970123022, 'number': 827}, 'overall_precision': 0.5452307692307692, 'overall_recall': 0.6060191518467852, 'overall_f1': 0.5740200842241658, 'overall_accuracy': 0.7245824487812365}
			------------EPOCH 20---------------
Loss:  tensor(80.2393, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(59.8618, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.8909, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.6506, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.0983, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.2129, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.6980, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.1878, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.9474, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.3035, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.9376, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.6178, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.8641, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.7077, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.8935, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.1620, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.6465, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.0119, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(71.3907, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.7482, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.0932, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.0056, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.5470, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.4970, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.8477, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.4417, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.9940, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(106.1151, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.5444, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.1929, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.9858, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.0867, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.9428, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.2232, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.3467, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.4693, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.2879, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.48620236530880423, 'recall': 0.5826771653543307, 'f1': 0.5300859598853869, 'number': 635}, 'P': {'precision': 0.5933562428407789, 'recall': 0.626360338573156, 'f1': 0.6094117647058824, 'number': 827}, 'overall_precision': 0.543451652386781, 'overall_recall': 0.6073871409028728, 'overall_f1': 0.5736434108527132, 'overall_accuracy': 0.7241476328107016}
			------------EPOCH 21---------------
Loss:  tensor(74.0900, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.7643, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.0628, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.5740, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.9800, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.3953, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.2586, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.8545, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.6764, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.6402, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.4099, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.8744, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.6947, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.0337, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.6086, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.6099, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.0124, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.3330, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.2472, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.7066, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.6357, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.3790, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.4912, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.4845, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.1341, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.2455, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.6535, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(96.4163, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.7702, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.7449, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.3252, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.1737, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.3010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.3835, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.9407, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.1182, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.8116, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4660691421254802, 'recall': 0.573228346456693, 'f1': 0.5141242937853108, 'number': 635}, 'P': {'precision': 0.5909090909090909, 'recall': 0.6287787182587666, 'f1': 0.6092560046865846, 'number': 827}, 'overall_precision': 0.5322095123419627, 'overall_recall': 0.6046511627906976, 'overall_f1': 0.5661223182837015, 'overall_accuracy': 0.7246336036012994}
			------------EPOCH 22---------------
Loss:  tensor(71.1808, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.7208, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.2864, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.7656, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.1519, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.0044, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.0470, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.9155, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.6168, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.3139, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.5318, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.7256, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.7920, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.9439, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.6451, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.9237, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.4994, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.6227, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.3356, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.4726, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.0995, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.2039, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.9410, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.8187, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.5235, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.8310, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.6449, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(83.7366, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.7123, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.1848, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.6688, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.5209, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.3109, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.4800, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.5673, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.9987, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.3725, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.46954314720812185, 'recall': 0.5826771653543307, 'f1': 0.5200281096275474, 'number': 635}, 'P': {'precision': 0.5904977375565611, 'recall': 0.6311970979443773, 'f1': 0.6101694915254237, 'number': 827}, 'overall_precision': 0.5334928229665071, 'overall_recall': 0.6101231190150479, 'overall_f1': 0.5692405871091257, 'overall_accuracy': 0.7253241936721487}
			------------EPOCH 23---------------
Loss:  tensor(68.5542, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.8422, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.9621, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.0771, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.2754, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.7107, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.1349, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.9562, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.7534, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.2593, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.3414, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.6982, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.1430, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.9406, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.0764, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.1414, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.4324, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.4314, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.8929, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.0133, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.9368, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.5310, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.9258, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.7317, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.1747, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.4356, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.0827, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(73.2108, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.7744, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.6592, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.5439, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.1091, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.5075, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.6837, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.7504, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.0561, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.2142, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.47414880201765447, 'recall': 0.5921259842519685, 'f1': 0.5266106442577031, 'number': 635}, 'P': {'precision': 0.5899772209567198, 'recall': 0.626360338573156, 'f1': 0.6076246334310851, 'number': 827}, 'overall_precision': 0.5350089766606823, 'overall_recall': 0.6114911080711354, 'overall_f1': 0.5706990105330354, 'overall_accuracy': 0.7248893777016139}
			------------EPOCH 24---------------
Loss:  tensor(65.7261, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.2832, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.4870, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.5069, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.6170, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.7075, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.3626, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.6332, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.9125, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.8655, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.3301, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.3671, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.7208, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.1986, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.3874, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.1224, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.6406, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.2989, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.1775, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.2529, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.8589, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.4595, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.5413, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.9326, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.9420, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.9118, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.5622, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(70.0635, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.2707, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.8332, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.2388, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.0176, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.5161, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.9782, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.5591, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.9886, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.5686, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4796954314720812, 'recall': 0.5952755905511811, 'f1': 0.5312719606465215, 'number': 635}, 'P': {'precision': 0.5838020247469067, 'recall': 0.6275695284159613, 'f1': 0.6048951048951049, 'number': 827}, 'overall_precision': 0.5348837209302325, 'overall_recall': 0.6135430916552668, 'overall_f1': 0.5715195922268238, 'overall_accuracy': 0.7241476328107016}
			------------EPOCH 25---------------
Loss:  tensor(64.7612, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.2230, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.3288, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.4689, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.3449, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.7079, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.5454, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.6785, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.7238, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.0889, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.8499, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.5783, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.0949, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.5434, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.2380, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.4038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.2872, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.0774, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.9861, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.8578, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.1516, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.9489, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.0659, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.5068, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.8738, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.0741, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.4309, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(65.8052, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.0619, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.8442, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.2170, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.3745, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.3919, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.3314, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.0575, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.7834, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.4080, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.48086734693877553, 'recall': 0.5937007874015748, 'f1': 0.5313601127554617, 'number': 635}, 'P': {'precision': 0.5813424345847554, 'recall': 0.6178960096735188, 'f1': 0.5990621336459554, 'number': 827}, 'overall_precision': 0.533974744437763, 'overall_recall': 0.6073871409028728, 'overall_f1': 0.56832, 'overall_accuracy': 0.7265519093536589}
			------------EPOCH 26---------------
Loss:  tensor(62.0391, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.6336, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.8636, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.9865, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.5996, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.9042, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(59.3540, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.6159, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.1147, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.0151, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.2284, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.8501, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.9921, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.0640, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.2559, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.8908, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.5349, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.2680, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.6063, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.6560, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.3997, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.5296, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.5610, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.0065, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.4680, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.5023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.1796, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(62.1477, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.8332, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.2145, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.9995, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.1265, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.7812, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.1421, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.7106, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.8991, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.2170, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.47848101265822784, 'recall': 0.5952755905511811, 'f1': 0.5305263157894736, 'number': 635}, 'P': {'precision': 0.5916289592760181, 'recall': 0.6324062877871826, 'f1': 0.6113383985973114, 'number': 827}, 'overall_precision': 0.538231780167264, 'overall_recall': 0.6162790697674418, 'overall_f1': 0.5746173469387755, 'overall_accuracy': 0.7257590096426836}
			------------EPOCH 27---------------
Loss:  tensor(63.2695, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.4096, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.4376, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.8877, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.1398, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.5752, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.6760, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.1922, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.8778, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.9206, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.2482, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.0770, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.1121, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.5070, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.3000, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.0425, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.4819, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.0539, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.7022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.4573, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.1430, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.2770, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.9119, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.2459, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.5448, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.8031, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.9627, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.9887, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.6899, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.9990, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.8233, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.3955, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.8789, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.6970, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.6896, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.1356, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.3576, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.48553459119496856, 'recall': 0.6078740157480315, 'f1': 0.5398601398601398, 'number': 635}, 'P': {'precision': 0.5888382687927107, 'recall': 0.6251511487303507, 'f1': 0.6064516129032258, 'number': 827}, 'overall_precision': 0.5397489539748954, 'overall_recall': 0.6176470588235294, 'overall_f1': 0.5760765550239235, 'overall_accuracy': 0.7262961352533442}
			------------EPOCH 28---------------
Loss:  tensor(66.5245, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.3064, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.3071, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.8222, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.5172, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.2560, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.4361, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.3684, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.9414, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.8475, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.7664, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.9102, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.3286, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.7097, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.6697, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.2407, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.3304, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.6309, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.8758, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.1048, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.4534, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.8790, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.2708, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.1525, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.2759, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.4232, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.3057, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.5268, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.4021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.9852, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.1573, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.8282, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.5589, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.4090, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.0694, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.6353, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.5437, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4840357598978289, 'recall': 0.5968503937007874, 'f1': 0.534555712270804, 'number': 635}, 'P': {'precision': 0.5853658536585366, 'recall': 0.6384522370012092, 'f1': 0.6107576633892423, 'number': 827}, 'overall_precision': 0.5382789317507418, 'overall_recall': 0.6203830369357045, 'overall_f1': 0.5764219891960597, 'overall_accuracy': 0.7251451518019285}
			------------EPOCH 29---------------
Loss:  tensor(65.1625, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.5313, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.0756, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.2222, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.0121, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.7365, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.1951, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.3193, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.0289, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.0135, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.2653, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.9329, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.4018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.7901, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.6817, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.8828, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.6701, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.2911, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.2175, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.5209, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.9043, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.4375, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.0651, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.6721, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.2420, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.4349, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.6484, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.4409, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.1916, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.5644, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.8402, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.3351, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.2411, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.8036, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.0429, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.1059, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.2034, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4830188679245283, 'recall': 0.6047244094488189, 'f1': 0.5370629370629372, 'number': 635}, 'P': {'precision': 0.5871040723981901, 'recall': 0.6275695284159613, 'f1': 0.6066627703097603, 'number': 827}, 'overall_precision': 0.5378201310303752, 'overall_recall': 0.6176470588235294, 'overall_f1': 0.5749761222540593, 'overall_accuracy': 0.7260147837429982}
			------------EPOCH 30---------------
Loss:  tensor(53.6007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.9772, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.5789, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.9272, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.6810, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.9893, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.5788, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.0210, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.2300, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.6582, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.6459, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.1812, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.1135, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.6305, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.4181, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.4273, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.2537, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.8533, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.9094, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.9397, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.2992, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.6259, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.3991, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.2385, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.0477, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.9588, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.0735, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.7705, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.1216, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.6065, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.1617, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.3765, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.0085, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.4592, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.2779, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.8700, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.1301, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.479539641943734, 'recall': 0.5905511811023622, 'f1': 0.5292872265349329, 'number': 635}, 'P': {'precision': 0.5837053571428571, 'recall': 0.6324062877871826, 'f1': 0.6070806732443412, 'number': 827}, 'overall_precision': 0.5351609058402861, 'overall_recall': 0.6142270861833106, 'overall_f1': 0.5719745222929937, 'overall_accuracy': 0.728009821725452}


		-------------RUN 4-----------
			------------EPOCH 1---------------
Loss:  tensor(1470.6606, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1591.3804, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1748.3250, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2422.8210, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1009.8492, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2773.5266, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3029.6201, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2446.9297, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1614.8585, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2720.4885, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1563.9320, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(778.9088, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2020.9318, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1042.6875, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2150.7146, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1391.3018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1431.5520, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2813.2134, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2552.3594, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1115.9141, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1453.6729, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1754.3088, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1813.1135, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1429.0953, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1842.5311, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1358.4182, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1884.3032, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1514.0082, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1327.7441, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2715.6562, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2119.0044, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(991.5768, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1822.5442, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1570.1377, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1309.3365, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2159.9424, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.014184397163120567, 'recall': 0.002652519893899204, 'f1': 0.00446927374301676, 'number': 754}, 'P': {'precision': 0.048205128205128206, 'recall': 0.049214659685863874, 'f1': 0.04870466321243524, 'number': 955}, 'overall_precision': 0.04390681003584229, 'overall_recall': 0.028671737858396724, 'overall_f1': 0.03469026548672567, 'overall_accuracy': 0.498038135884904}
			------------EPOCH 2---------------
Loss:  tensor(1101.7213, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1193.5447, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1435.2288, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1925.5548, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(800.2172, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2072.3079, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2089.0020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1823.5818, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1254.4152, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2030.7141, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1203.3491, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(690.4940, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1647.2358, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(897.2328, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1785.2235, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1157.0983, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1165.4459, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2187.0581, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2190.1023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(963.8407, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1256.2424, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1540.4207, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1498.3351, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1173.5863, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1521.5333, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1118.3086, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1463.6434, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1222.1677, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1119.8198, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2231.8540, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1795.6752, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(834.3160, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1541.8014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1336.6307, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1160.3164, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1908.1715, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.0773286467486819, 'recall': 0.058355437665782495, 'f1': 0.06651549508692366, 'number': 754}, 'P': {'precision': 0.18628719275549807, 'recall': 0.15078534031413612, 'f1': 0.16666666666666666, 'number': 955}, 'overall_precision': 0.14008941877794337, 'overall_recall': 0.11000585137507314, 'overall_f1': 0.12323828253031792, 'overall_accuracy': 0.5096028085633648}
			------------EPOCH 3---------------
Loss:  tensor(922.2186, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(971.5890, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1172.0303, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1598.4905, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(733.4637, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1754.5734, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1844.4232, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1593.7841, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1012.2219, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1736.6299, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1023.7791, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(528.3491, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1294.9978, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(683.9252, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1433.4146, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1009.0975, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(984.7910, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1758.8146, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1761.8726, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(786.2113, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(952.9991, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1309.8857, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1260.6240, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(965.5292, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1228.2947, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(904.1796, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1182.8167, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(996.7415, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(922.0893, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1863.2861, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1580.9475, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(714.5204, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1250.7067, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1110.7458, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1036.1470, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1677.0862, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.17136659436008678, 'recall': 0.20954907161803712, 'f1': 0.18854415274463007, 'number': 754}, 'P': {'precision': 0.3347763347763348, 'recall': 0.24293193717277486, 'f1': 0.2815533980582524, 'number': 955}, 'overall_precision': 0.24148606811145512, 'overall_recall': 0.22820362785254536, 'overall_f1': 0.23465703971119137, 'overall_accuracy': 0.5551501801243661}
			------------EPOCH 4---------------
Loss:  tensor(803.8807, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(816.9439, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(939.2635, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1357.5400, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(703.7396, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1512.0720, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1531.1925, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1369.5912, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(804.7487, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1493.6531, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(882.9606, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(406.1457, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1000.0261, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(515.7527, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1136.2174, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(881.6209, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(759.9257, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1254.2134, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1274.6456, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(639.3719, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(666.1494, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1052.3783, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1065.1881, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(842.1261, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(982.9617, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(772.9898, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(987.5394, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(835.6177, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(747.7319, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1581.7473, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1468.1934, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(589.7373, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1020.2193, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(928.1914, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(992.5824, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1442.5043, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.2270967741935484, 'recall': 0.46684350132625996, 'f1': 0.3055555555555556, 'number': 754}, 'P': {'precision': 0.42317380352644834, 'recall': 0.17591623036649215, 'f1': 0.2485207100591716, 'number': 955}, 'overall_precision': 0.2670775552131484, 'overall_recall': 0.3042715038033938, 'overall_f1': 0.2844638949671772, 'overall_accuracy': 0.5289231545857139}
			------------EPOCH 5---------------
Loss:  tensor(738.4169, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(681.8185, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(749.8223, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1103.3698, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(630.3129, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1341.2778, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1432.3876, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1189.3708, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(696.6896, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1341.2062, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(841.3500, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(331.7443, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(825.9182, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(424.2688, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(967.3814, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(746.6575, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(626.0721, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(993.9593, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(907.2502, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(474.7588, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(427.4172, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(868.7643, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(894.2548, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(694.7086, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(790.4312, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(692.4540, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(823.6191, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(715.3248, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(705.5590, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1485.9243, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1472.4482, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(557.8175, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(902.8262, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(865.4642, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1102.8669, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1395.3353, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.31101321585903086, 'recall': 0.46816976127320953, 'f1': 0.3737427210164108, 'number': 754}, 'P': {'precision': 0.4948770491803279, 'recall': 0.5057591623036649, 'f1': 0.5002589331952356, 'number': 955}, 'overall_precision': 0.3960208432022738, 'overall_recall': 0.489174956114687, 'overall_f1': 0.43769633507853406, 'overall_accuracy': 0.598196461760859}
			------------EPOCH 6---------------
Loss:  tensor(691.9576, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(669.2523, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(717.1602, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1111.4648, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(454.8840, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1128.2604, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1092.9099, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(972.8096, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(526.9807, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1194.6904, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(748.2463, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(284.0757, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(763.3515, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(348.7315, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(844.3198, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(701.2427, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(572.6667, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1091.4548, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(864.2135, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(386.6920, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(346.8704, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(833.2254, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(915.4530, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(732.9240, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(679.1776, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(550.8352, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(744.5320, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(640.9688, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(586.6719, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1161.9092, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1117.4677, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(448.4829, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(770.6623, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(700.9373, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(886.1466, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1076.1226, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4173387096774194, 'recall': 0.27453580901856767, 'f1': 0.33120000000000005, 'number': 754}, 'P': {'precision': 0.40699815837937386, 'recall': 0.694240837696335, 'f1': 0.5131578947368421, 'number': 955}, 'overall_precision': 0.40941176470588236, 'overall_recall': 0.5090696313633704, 'overall_f1': 0.4538341158059468, 'overall_accuracy': 0.6138454831233795}
			------------EPOCH 7---------------
Loss:  tensor(719.7318, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(711.5725, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(817.9347, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1350.6018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(377.8268, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1102.2058, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(976.1835, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1089.9951, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(550.9880, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1386.1226, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(773.9651, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(275.1932, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(762.4608, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(318.2704, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(863.7675, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(780.0321, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(789.5850, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1354.5278, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1063.0416, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(338.6264, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(259.8464, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(663.1183, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(747.2772, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(511.6492, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(609.3296, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(553.1224, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(543.3893, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(514.9377, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(533.9539, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1036.2983, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(993.3229, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(438.3513, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(837.4180, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(687.6772, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(763.8660, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1066.0609, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3271276595744681, 'recall': 0.16312997347480107, 'f1': 0.2176991150442478, 'number': 754}, 'P': {'precision': 0.4717906786590352, 'recall': 0.6041884816753926, 'f1': 0.5298438934802572, 'number': 955}, 'overall_precision': 0.4377736085053158, 'overall_recall': 0.4095962551199532, 'overall_f1': 0.42321644498186217, 'overall_accuracy': 0.6333264495995962}
			------------EPOCH 8---------------
Loss:  tensor(573.5144, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(745.9365, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(695.0909, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1184.0562, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(415.5341, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(923.3724, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(885.8658, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(983.7405, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(396.0245, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(875.8848, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(669.3659, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(265.8649, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(495.9686, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(268.7467, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(626.7119, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(499.9467, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(423.3678, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(612.6830, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(572.0432, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(283.4082, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(166.9353, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(475.1776, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(513.7134, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(348.6846, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(483.3625, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(492.1349, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(384.8786, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(414.6375, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(521.4197, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1002.4899, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1062.4675, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(468.7593, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(748.7371, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(641.5679, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(861.6099, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1091.8958, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.40551724137931033, 'recall': 0.38992042440318303, 'f1': 0.3975659229208925, 'number': 754}, 'P': {'precision': 0.45173176123802505, 'recall': 0.6418848167539267, 'f1': 0.5302768166089965, 'number': 955}, 'overall_precision': 0.4356388088376561, 'overall_recall': 0.5307197191339965, 'overall_f1': 0.47850171458718016, 'overall_accuracy': 0.6486542300543815}
			------------EPOCH 9---------------
Loss:  tensor(467.1840, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(455.8410, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(528.0684, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(818.8677, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(276.5205, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(708.4531, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(637.6224, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(790.8215, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(246.4564, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(799.0762, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(517.5673, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(172.6945, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(453.3073, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(222.6345, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(546.2839, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(396.9410, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(349.3714, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(520.6345, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(621.9202, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(308.7343, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(150.1007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(492.2681, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(481.3834, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(384.4006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(356.5977, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(339.1375, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(389.4115, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(351.0245, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(349.0195, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(684.0211, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(629.9298, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(275.8872, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(525.6367, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(487.1252, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(499.5905, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(687.9205, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3159806295399516, 'recall': 0.6923076923076923, 'f1': 0.4339152119700748, 'number': 754}, 'P': {'precision': 0.5955334987593052, 'recall': 0.2513089005235602, 'f1': 0.35346097201767307, 'number': 955}, 'overall_precision': 0.3708029197080292, 'overall_recall': 0.44587478057343477, 'overall_f1': 0.40488841657810837, 'overall_accuracy': 0.556503981092678}
			------------EPOCH 10---------------
Loss:  tensor(610.4886, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(523.5577, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(539.7172, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(870.9249, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(412.3321, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(859.6663, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(824.0063, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1097.1382, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(369.4976, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(810.6592, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(547.1853, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(154.7535, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(314.4036, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(169.1154, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(422.1426, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(387.5305, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(292.6589, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(431.0327, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(474.7263, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(213.9967, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(142.9037, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(441.9828, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(419.7754, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(331.4409, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(425.8605, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(471.1617, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(421.1223, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(410.7369, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(494.1024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1021.7888, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1025.8075, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(422.0586, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(567.4578, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(508.1706, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(663.7957, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(840.9962, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3488943488943489, 'recall': 0.376657824933687, 'f1': 0.36224489795918363, 'number': 754}, 'P': {'precision': 0.48811700182815354, 'recall': 0.5591623036649215, 'f1': 0.5212298682284041, 'number': 955}, 'overall_precision': 0.42872117400419285, 'overall_recall': 0.478642480983031, 'overall_f1': 0.45230854299142936, 'overall_accuracy': 0.65179780179436}
			------------EPOCH 11---------------
Loss:  tensor(292.2880, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(413.1501, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(335.0612, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(583.2200, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(224.2475, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(433.3525, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(513.6434, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(614.7595, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(191.1076, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(522.4299, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(386.8766, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(122.7036, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(366.0417, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(181.8594, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(455.8297, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(384.2956, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(301.3036, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(589.2300, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(499.2657, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(198.2672, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(104.5245, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(286.1254, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(355.5843, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(231.9399, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(264.7168, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(287.9515, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(218.5106, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(231.0492, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(200.1579, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(415.8318, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(443.6537, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(199.3463, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(345.6888, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(295.7507, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(330.2203, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(370.2488, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.42473118279569894, 'recall': 0.41909814323607425, 'f1': 0.4218958611481976, 'number': 754}, 'P': {'precision': 0.46996996996997, 'recall': 0.6554973821989529, 'f1': 0.5474420638390904, 'number': 955}, 'overall_precision': 0.45375722543352603, 'overall_recall': 0.5511995318899942, 'overall_f1': 0.49775429326287973, 'overall_accuracy': 0.6556756384663042}
			------------EPOCH 12---------------
Loss:  tensor(274.5773, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(341.8852, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(334.3604, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(558.0107, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(170.2964, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(400.6264, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(376.7748, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(630.5563, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(163.6859, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(466.9465, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(271.1201, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(87.2650, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(188.4902, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(103.1306, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(271.7507, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(173.9029, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(149.5348, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(226.9525, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(254.8095, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(135.5434, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(65.2561, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(186.8194, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(175.0275, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(122.1045, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(153.4595, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(166.6996, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(126.7527, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(157.6838, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(125.6409, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(320.0013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(312.3643, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(150.5760, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(201.9895, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(285.7566, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(234.6636, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(244.1487, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3883129123468426, 'recall': 0.5464190981432361, 'f1': 0.4539944903581268, 'number': 754}, 'P': {'precision': 0.5592255125284739, 'recall': 0.5141361256544502, 'f1': 0.5357337697763229, 'number': 955}, 'overall_precision': 0.4657039711191336, 'overall_recall': 0.5283791691047396, 'overall_f1': 0.4950657894736842, 'overall_accuracy': 0.6551478855464538}
			------------EPOCH 13---------------
Loss:  tensor(183.1597, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(232.7092, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(180.1607, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(293.1520, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(98.3005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(206.4915, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(275.4976, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(411.2256, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(88.3536, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(247.0289, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(154.1444, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(70.4178, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(155.5627, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(67.4202, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(151.4750, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(151.0894, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(103.7019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(185.7342, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(194.3239, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(90.2950, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.9276, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(135.5295, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(118.1636, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(80.6644, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(107.0192, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(128.5525, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(82.5543, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(88.1122, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(82.8597, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(172.0966, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(220.7483, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(117.2009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(140.4648, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(181.8872, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(160.6506, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(134.6411, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.40168539325842695, 'recall': 0.3793103448275862, 'f1': 0.3901773533424284, 'number': 754}, 'P': {'precision': 0.49341021416803954, 'recall': 0.6272251308900524, 'f1': 0.5523282618718303, 'number': 955}, 'overall_precision': 0.45950155763239875, 'overall_recall': 0.5178466939730837, 'overall_f1': 0.48693259972489683, 'overall_accuracy': 0.6735045088455979}
			------------EPOCH 14---------------
Loss:  tensor(134.9347, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(183.5548, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(149.8798, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(199.1039, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(67.8447, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(148.1564, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(198.4457, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(329.2825, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(67.2213, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(126.2528, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(90.2184, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.6922, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(97.0851, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.5360, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(103.3173, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(100.4229, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.4580, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(142.3643, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(145.8243, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.0679, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.6811, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(94.5770, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(69.6988, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.2054, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(71.9232, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(97.0735, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(62.3759, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.5178, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(70.2766, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(124.5348, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(157.2560, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(81.6401, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(107.6372, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(154.9739, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(131.9773, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(102.9121, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4037698412698413, 'recall': 0.5397877984084881, 'f1': 0.4619750283768445, 'number': 754}, 'P': {'precision': 0.5358948432760364, 'recall': 0.5549738219895288, 'f1': 0.5452674897119342, 'number': 955}, 'overall_precision': 0.46920380570856285, 'overall_recall': 0.5482738443534231, 'overall_f1': 0.5056664867781975, 'overall_accuracy': 0.6664601546545513}
			------------EPOCH 15---------------
Loss:  tensor(95.7456, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(96.9809, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(93.6654, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(120.1037, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.2070, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(114.5569, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(159.4117, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(190.4662, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.1965, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(75.5618, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.2293, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.0298, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.1787, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.9598, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(64.2565, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(71.0528, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.8104, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(120.4078, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(101.0084, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.4403, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.9657, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(73.0988, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.4201, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.9604, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.8262, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(84.2263, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.2260, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.3591, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.7984, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(90.9816, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(112.4202, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.6335, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(76.8630, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(122.2181, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(104.7156, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(72.2691, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4230769230769231, 'recall': 0.46684350132625996, 'f1': 0.4438839848675914, 'number': 754}, 'P': {'precision': 0.5067996373526745, 'recall': 0.5853403141361256, 'f1': 0.543245869776482, 'number': 955}, 'overall_precision': 0.47080103359173125, 'overall_recall': 0.5330602691632533, 'overall_f1': 0.5, 'overall_accuracy': 0.6761891650030977}
			------------EPOCH 16---------------
Loss:  tensor(73.6930, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(61.7435, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(64.7173, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(89.3614, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.8051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(95.9158, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(132.9032, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(125.9485, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.3851, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.3587, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.2063, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.7817, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.0058, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.8772, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.7374, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.5721, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.5788, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(103.3222, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(80.4024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.8670, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.8982, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.6816, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.8048, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.9703, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.0015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.4142, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.1177, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.4981, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.6249, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(65.2606, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(80.6823, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.2093, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(61.7265, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(91.0256, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(98.6256, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(61.2246, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4118895966029724, 'recall': 0.5145888594164456, 'f1': 0.45754716981132076, 'number': 754}, 'P': {'precision': 0.5146198830409356, 'recall': 0.5528795811518324, 'f1': 0.5330641090358406, 'number': 955}, 'overall_precision': 0.4654471544715447, 'overall_recall': 0.5359859566998244, 'overall_f1': 0.49823225455534403, 'overall_accuracy': 0.6740322617654483}
			------------EPOCH 17---------------
Loss:  tensor(60.2645, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.3495, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.4436, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(66.8009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.4716, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(77.7446, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(109.7547, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(90.4688, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.5266, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.9394, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.7653, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.7557, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.2227, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.9937, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.6300, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.9995, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.1579, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(91.3802, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(67.9776, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.7716, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.2367, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.3123, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.0117, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.6888, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.9162, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.2308, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.1172, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.3977, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.6980, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.4980, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.5818, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.8369, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.9912, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(78.7660, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(78.1461, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.7086, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4134275618374558, 'recall': 0.46551724137931033, 'f1': 0.4379288833437305, 'number': 754}, 'P': {'precision': 0.5054054054054054, 'recall': 0.587434554973822, 'f1': 0.5433414043583535, 'number': 955}, 'overall_precision': 0.4655436447166922, 'overall_recall': 0.5336454066705676, 'overall_f1': 0.49727371864776443, 'overall_accuracy': 0.6752942796172644}
			------------EPOCH 18---------------
Loss:  tensor(49.3370, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.2687, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.7105, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.5058, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.1214, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(70.6554, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(98.0034, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(71.1347, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.9936, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.5848, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.9050, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.2683, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.3740, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.3288, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.1746, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.4750, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.3984, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(77.5299, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.7980, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.4522, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.7890, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.4696, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.0608, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.8127, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.7857, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.0389, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.5768, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.9036, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.1328, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.1057, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.3317, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.6283, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.8938, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(69.1543, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(67.8780, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.7858, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3950617283950617, 'recall': 0.5092838196286472, 'f1': 0.44495944380069524, 'number': 754}, 'P': {'precision': 0.5093046033300686, 'recall': 0.5445026178010471, 'f1': 0.5263157894736843, 'number': 955}, 'overall_precision': 0.4535875564475665, 'overall_recall': 0.5289643066120538, 'overall_f1': 0.48838465694219335, 'overall_accuracy': 0.6649686790114958}
			------------EPOCH 19---------------
Loss:  tensor(42.6733, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.0376, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.7557, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.1436, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.8889, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(79.3253, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(100.4343, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(72.1053, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.8986, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.3461, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.0572, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.4963, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.3854, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.3828, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.6017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.6097, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.3403, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(85.2035, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(74.1094, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.9463, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.3207, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.8915, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.5183, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.0860, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.4996, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.8835, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.4335, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.8951, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.7469, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(61.3082, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(62.1741, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.5840, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(62.7175, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(73.4844, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(64.4212, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.0204, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.389010989010989, 'recall': 0.46949602122015915, 'f1': 0.4254807692307692, 'number': 754}, 'P': {'precision': 0.4867094408799267, 'recall': 0.556020942408377, 'f1': 0.5190615835777126, 'number': 955}, 'overall_precision': 0.44227886056971516, 'overall_recall': 0.5178466939730837, 'overall_f1': 0.477088948787062, 'overall_accuracy': 0.6719212500860466}
			------------EPOCH 20---------------
Loss:  tensor(43.8348, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.0741, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.7385, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.6618, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.6636, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(83.9708, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(100.7936, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(78.9564, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.3595, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.8599, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.8346, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.1249, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.7230, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.2513, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.7645, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.5825, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.2367, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(91.9725, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(89.5783, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.1019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.2677, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.7400, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.0467, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.4416, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.1961, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.0642, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.3219, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.6217, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.6546, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.5124, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.1344, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.6922, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.2010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(59.6116, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.0116, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.3040, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4059003051881994, 'recall': 0.5291777188328912, 'f1': 0.459412780656304, 'number': 754}, 'P': {'precision': 0.4951644100580271, 'recall': 0.5361256544502618, 'f1': 0.5148315736551031, 'number': 955}, 'overall_precision': 0.4516608824987605, 'overall_recall': 0.5330602691632533, 'overall_f1': 0.48899624261943103, 'overall_accuracy': 0.6670108533535256}
			------------EPOCH 21---------------
Loss:  tensor(37.1114, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.0349, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.4794, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.2117, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.2854, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(72.0945, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(90.9073, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(74.4815, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.2448, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.1973, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.0205, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.6402, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.7575, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.5790, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.0577, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.1306, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.8557, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(68.9323, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.8447, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.6539, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.1397, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.2005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.0865, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.3700, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.4431, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.8722, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.5963, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.8499, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.3476, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.4881, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.4365, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.2727, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.2289, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.0493, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.6270, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.9777, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4239733629300777, 'recall': 0.506631299734748, 'f1': 0.461631419939577, 'number': 754}, 'P': {'precision': 0.5018315018315018, 'recall': 0.5738219895287958, 'f1': 0.5354176844162188, 'number': 955}, 'overall_precision': 0.46663321625689913, 'overall_recall': 0.5441778818022235, 'overall_f1': 0.5024311183144247, 'overall_accuracy': 0.6799981643376701}
			------------EPOCH 22---------------
Loss:  tensor(33.3561, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.4173, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.8197, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.2610, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.6995, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(66.6755, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(84.7637, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.5217, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.6241, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.8688, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.6632, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.9627, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.8055, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.7520, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.5076, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.7400, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.2696, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(61.7146, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.3329, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.0785, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.8781, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.4881, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.6138, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.2553, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.8016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.1456, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.9257, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.7236, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.7045, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.0058, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.4336, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.6225, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.9868, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.4151, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.2588, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.6364, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4196035242290749, 'recall': 0.5053050397877984, 'f1': 0.4584837545126354, 'number': 754}, 'P': {'precision': 0.5022914757103575, 'recall': 0.5738219895287958, 'f1': 0.5356793743890518, 'number': 955}, 'overall_precision': 0.46473236618309155, 'overall_recall': 0.5435927442949093, 'overall_f1': 0.5010787486515642, 'overall_accuracy': 0.6800899474541658}
			------------EPOCH 23---------------
Loss:  tensor(29.3361, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.3267, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.1977, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.5276, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.5531, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(61.2104, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(77.6353, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.3171, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.5151, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.1809, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.2328, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.1050, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.4322, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.4105, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.7919, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.6301, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.0413, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.1178, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.8365, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.0137, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.7361, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.0495, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.0464, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.5532, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.2793, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.0755, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.3998, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.8442, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.6846, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.7823, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.7720, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.1057, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.9579, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.2618, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.9735, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.5085, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.42857142857142855, 'recall': 0.5053050397877984, 'f1': 0.4637857577601948, 'number': 754}, 'P': {'precision': 0.5050878815911193, 'recall': 0.5717277486910994, 'f1': 0.5363457760314341, 'number': 955}, 'overall_precision': 0.47055837563451774, 'overall_recall': 0.5424224692802808, 'overall_f1': 0.5039412883935851, 'overall_accuracy': 0.6789885500562172}
			------------EPOCH 24---------------
Loss:  tensor(26.6352, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.3051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.5113, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.2987, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.5455, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.0939, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(73.3798, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.1480, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.5383, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.4440, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.3465, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.8068, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.5762, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.3970, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.8628, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.1238, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.6952, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.2884, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.4263, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.0160, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.6328, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.1297, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.5529, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.6790, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.4190, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.0219, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.5276, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.3669, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.6172, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.4286, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.4044, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.2635, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.3681, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.8495, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.8758, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.4794, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.43829296424452135, 'recall': 0.5039787798408488, 'f1': 0.4688463911165947, 'number': 754}, 'P': {'precision': 0.501834862385321, 'recall': 0.5727748691099477, 'f1': 0.5349633251833741, 'number': 955}, 'overall_precision': 0.47368421052631576, 'overall_recall': 0.5424224692802808, 'overall_f1': 0.5057283142389526, 'overall_accuracy': 0.6799293270002983}
			------------EPOCH 25---------------
Loss:  tensor(24.4763, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.6521, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.1559, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.4080, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.4217, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.4122, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(69.8768, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.3810, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.2561, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.8729, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.8969, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.6866, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.6342, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.3817, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.5557, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.9830, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.4263, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.6399, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.1438, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.2491, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.4826, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.5679, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.1544, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.7638, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.3902, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.3062, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.1635, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.9487, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.8897, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.4212, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.2471, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.7960, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.9538, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.7370, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.2574, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.7218, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.427594070695553, 'recall': 0.4973474801061008, 'f1': 0.4598405885959534, 'number': 754}, 'P': {'precision': 0.5009191176470589, 'recall': 0.5706806282722513, 'f1': 0.5335291238374938, 'number': 955}, 'overall_precision': 0.4681933842239186, 'overall_recall': 0.5383265067290813, 'overall_f1': 0.5008165487207403, 'overall_accuracy': 0.6784607971363668}
			------------EPOCH 26---------------
Loss:  tensor(22.7737, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.5777, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.1359, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.9289, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.3710, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.0448, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(66.8045, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.7630, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.2452, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.4539, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.5249, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.6093, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.3075, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.6262, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.3093, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.0524, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.4307, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.9778, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.2070, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.5136, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.6357, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.3336, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.0578, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.4174, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.6856, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.4016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.3434, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.7383, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.5496, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.5684, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.4980, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.5462, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.2384, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.1973, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.5256, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.3368, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4261168384879725, 'recall': 0.493368700265252, 'f1': 0.45728334357713585, 'number': 754}, 'P': {'precision': 0.49817184643510054, 'recall': 0.5706806282722513, 'f1': 0.5319668130795511, 'number': 955}, 'overall_precision': 0.46619217081850534, 'overall_recall': 0.5365710942071387, 'overall_f1': 0.4989118607181719, 'overall_accuracy': 0.6777035864252771}
			------------EPOCH 27---------------
Loss:  tensor(21.5642, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.6351, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.3307, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.7077, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.5830, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.1120, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(64.1180, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.2532, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.6616, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.4237, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.6046, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.7159, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.0244, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.8268, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.1872, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.2730, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.6545, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.9624, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.6347, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.9743, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.0451, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.4972, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.2530, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.3629, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.1182, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.7129, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.6543, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.8207, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.5678, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.8362, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.9664, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.5569, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.0588, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.0300, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.2923, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.3084, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.42303306727480045, 'recall': 0.4920424403183024, 'f1': 0.4549356223175966, 'number': 754}, 'P': {'precision': 0.49501359927470534, 'recall': 0.5717277486910994, 'f1': 0.5306122448979591, 'number': 955}, 'overall_precision': 0.4631313131313131, 'overall_recall': 0.5365710942071387, 'overall_f1': 0.49715370018975336, 'overall_accuracy': 0.676418622794337}
			------------EPOCH 28---------------
Loss:  tensor(20.6568, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.1668, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.7707, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.8507, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.8947, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.4291, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(61.5743, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.0991, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.3793, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.6543, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.8272, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.0533, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.1370, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.3865, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.4038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.7039, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.0487, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.6395, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.2775, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.5603, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.6019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.8547, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.6288, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.5847, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.6669, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.6100, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.9894, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.1226, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.8035, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.8269, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.7762, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.6967, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.0323, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.6928, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.3135, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.3804, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4223245109321059, 'recall': 0.48673740053050396, 'f1': 0.452248921749846, 'number': 754}, 'P': {'precision': 0.4936708860759494, 'recall': 0.5717277486910994, 'f1': 0.529839883551674, 'number': 955}, 'overall_precision': 0.4622784810126582, 'overall_recall': 0.5342305441778818, 'overall_f1': 0.495656894679696, 'overall_accuracy': 0.6757302494206191}
			------------EPOCH 29---------------
Loss:  tensor(20.0260, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.5293, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.3280, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.1639, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.3858, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.9295, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(59.1944, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.0864, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.3529, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.0740, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.2053, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.3642, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.3488, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.0245, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.7184, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.2519, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.5733, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.1278, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.1447, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.1594, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.2806, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.3587, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.0758, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.9657, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.2896, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.0351, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.4298, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.5556, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.0994, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.8439, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.3995, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.6207, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.3098, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.8898, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.5251, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.6433, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.41316685584562995, 'recall': 0.4827586206896552, 'f1': 0.4452599388379205, 'number': 754}, 'P': {'precision': 0.48868778280542985, 'recall': 0.5654450261780105, 'f1': 0.5242718446601942, 'number': 955}, 'overall_precision': 0.4551863041289023, 'overall_recall': 0.5289643066120538, 'overall_f1': 0.48930987821380245, 'overall_accuracy': 0.6749959844886533}
			------------EPOCH 30---------------
Loss:  tensor(19.4815, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.1629, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.9728, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.5975, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.8923, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.5861, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.1557, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.1434, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.5582, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.6068, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.6977, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.9042, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.6602, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.7317, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.1656, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.8625, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.1721, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.3498, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.1796, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.8443, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.9938, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.9596, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.6165, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.4666, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.9558, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.1481, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.7634, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.0738, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.5135, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.1296, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.7180, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.1501, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.7531, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.4491, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.9357, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.0045, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.40408626560726446, 'recall': 0.47214854111405835, 'f1': 0.4354740061162079, 'number': 754}, 'P': {'precision': 0.48512173128944996, 'recall': 0.5633507853403141, 'f1': 0.5213178294573644, 'number': 955}, 'overall_precision': 0.4492462311557789, 'overall_recall': 0.5231129315389117, 'overall_f1': 0.48337388483373883, 'overall_accuracy': 0.6754778458502558}


		-------------RUN 5-----------
			------------EPOCH 1---------------
Loss:  tensor(1969.3932, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1949.7880, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2181.0825, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1789.6956, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1850.4329, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1546.8271, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1625.8921, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1904.0742, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1559.3362, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2975.9858, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2864.6680, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2042.7839, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1921.5645, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1072.4528, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(775.4520, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(820.8204, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1667.0593, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1512.2515, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1426.2458, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1594.3164, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1762.4133, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1013.6502, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1643.6899, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2282.1653, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2040.5825, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1816.2356, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(814.6380, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1436.4119, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1319.4694, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1450.8651, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1715.8986, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2997.0371, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2062.8186, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1978.5851, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1216.4862, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2058.5569, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1274.6345, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.09368635437881874, 'recall': 0.060209424083769635, 'f1': 0.07330677290836654, 'number': 764}, 'P': {'precision': 0.22256284488044145, 'recall': 0.37733887733887733, 'f1': 0.2799845738526803, 'number': 962}, 'overall_precision': 0.19274269557021678, 'overall_recall': 0.2369640787949015, 'overall_f1': 0.21257796257796258, 'overall_accuracy': 0.5453483660286511}
			------------EPOCH 2---------------
Loss:  tensor(1439.4019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1469.7670, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1479.6003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1205.6902, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1361.9917, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1195.0486, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1267.3884, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1496.0947, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1387.4409, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2470.1289, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2117.4392, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1622.3174, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1472.6250, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(898.4183, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(630.4989, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(706.9177, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1514.0146, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1347.6069, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1291.9983, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1324.8768, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1411.9170, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(874.4222, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1400.2295, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2048.5288, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1716.4860, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1512.6796, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(669.3891, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1134.7480, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1088.6401, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1271.3193, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1461.3638, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2446.5540, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1849.8662, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1765.3878, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1058.0238, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1767.6605, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1114.6976, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.14583333333333334, 'recall': 0.03664921465968586, 'f1': 0.05857740585774058, 'number': 764}, 'P': {'precision': 0.1734338747099768, 'recall': 0.3108108108108108, 'f1': 0.22263588979895757, 'number': 962}, 'overall_precision': 0.17066805845511482, 'overall_recall': 0.18945538818076477, 'overall_f1': 0.17957166392092258, 'overall_accuracy': 0.5785521893547542}
			------------EPOCH 3---------------
Loss:  tensor(1313.0354, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1315.8401, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1355.4207, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1097.8083, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1179.3506, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1118.4617, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1137.7329, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1278.2397, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1025.0479, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1766.0370, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1692.1094, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1343.4666, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1100.6915, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(789.0206, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(553.1569, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(582.5226, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1277.5684, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1101.9154, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1044.6487, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1220.2075, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1273.6448, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(791.6488, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1157.5322, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1699.4841, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1479.3633, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1395.0859, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(594.0073, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(994.6313, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(936.9944, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1134.9614, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1211.1201, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2093.9504, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1633.1451, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1468.4014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(816.8038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1423.0583, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(957.7620, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.23174603174603176, 'recall': 0.19109947643979058, 'f1': 0.2094691535150646, 'number': 764}, 'P': {'precision': 0.35782747603833864, 'recall': 0.4656964656964657, 'f1': 0.4046973803071364, 'number': 962}, 'overall_precision': 0.31562167906482463, 'overall_recall': 0.34414831981460026, 'overall_f1': 0.32926829268292684, 'overall_accuracy': 0.6146163563988273}
			------------EPOCH 4---------------
Loss:  tensor(1037.7217, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1227.6853, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1182.9707, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(983.5362, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1081.0061, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(776.1718, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(867.3413, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1102.3800, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(818.0242, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1397.2898, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1460.0439, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1131.7251, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(850.7433, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(670.9436, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(474.9977, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(485.2701, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1038.2357, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(836.9952, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(774.5649, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1141.6752, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1273.2985, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(743.6111, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1013.9277, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1486.8240, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1274.0737, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1331.5699, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(545.8411, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(898.1060, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(777.1938, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(914.4118, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(954.3710, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1815.6448, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1459.5442, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1241.3275, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(680.7997, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1252.8318, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(893.6335, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.22898154477101845, 'recall': 0.43848167539267013, 'f1': 0.3008531656937585, 'number': 764}, 'P': {'precision': 0.29256594724220625, 'recall': 0.12681912681912683, 'f1': 0.17693981145757795, 'number': 962}, 'overall_precision': 0.24308510638297873, 'overall_recall': 0.264774044032445, 'overall_f1': 0.25346644481419855, 'overall_accuracy': 0.5268276404547947}
			------------EPOCH 5---------------
Loss:  tensor(913.2155, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1222.5388, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1238.3035, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1064.8317, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1035.7966, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(645.6044, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(749.3992, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1133.4086, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(776.6279, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1318.6882, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1375.2424, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1114.9990, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(778.5155, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(552.5770, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(415.0265, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(408.1043, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(723.9666, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(581.9563, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(521.8227, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1012.7054, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1195.6582, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(724.0190, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(915.1140, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1344.5227, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1265.6317, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1421.2003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(606.2742, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(974.1149, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(756.4359, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(827.2974, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(920.3618, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2052.6755, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1372.0110, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1270.5526, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(611.5764, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1257.5588, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(884.9131, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3098814229249012, 'recall': 0.5130890052356021, 'f1': 0.38639724001971415, 'number': 764}, 'P': {'precision': 0.4389086595492289, 'recall': 0.38461538461538464, 'f1': 0.4099722991689751, 'number': 962}, 'overall_precision': 0.36148007590132825, 'overall_recall': 0.44148319814600234, 'overall_f1': 0.3974960876369327, 'overall_accuracy': 0.600195456820728}
			------------EPOCH 6---------------
Loss:  tensor(792.3455, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1069.5488, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(950.7767, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(808.5760, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(926.9248, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(533.6983, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(616.0497, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1019.1605, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(736.5923, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1262.1584, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1382.8862, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1135.5320, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(987.7234, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(544.4779, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(459.5966, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(436.4624, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(544.5579, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(503.7476, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(484.3709, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1027.1627, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(978.1379, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(547.5608, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(860.2520, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1248.5007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1149.9978, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1133.1229, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(398.9618, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(676.5117, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(654.6558, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(719.6550, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(781.6489, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1338.5281, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1148.7152, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1032.1777, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(517.2531, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1061.4497, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(837.1865, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4134078212290503, 'recall': 0.193717277486911, 'f1': 0.2638146167557932, 'number': 764}, 'P': {'precision': 0.3978685612788632, 'recall': 0.6985446985446986, 'f1': 0.5069784986797435, 'number': 962}, 'overall_precision': 0.40058622374206154, 'overall_recall': 0.47508690614136734, 'overall_f1': 0.4346673734428837, 'overall_accuracy': 0.6440778966939193}
			------------EPOCH 7---------------
Loss:  tensor(818.6530, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1078.2108, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(860.2958, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(689.0134, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(846.3660, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(540.9412, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(595.4609, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(930.2358, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(763.9814, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1227.7341, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1051.6405, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(928.2688, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(563.2353, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(402.2161, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(274.8535, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(322.6626, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(388.6111, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(325.2977, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(291.7563, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(720.6643, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(755.6180, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(417.8074, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(572.2642, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(842.9606, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(841.1572, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(945.5300, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(329.8610, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(547.1475, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(605.0591, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(633.2211, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(711.7301, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1311.0229, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1169.9189, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(857.5565, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(457.2032, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(934.4935, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(657.5438, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.38192090395480227, 'recall': 0.4424083769633508, 'f1': 0.4099454214675561, 'number': 764}, 'P': {'precision': 0.48408239700374533, 'recall': 0.5374220374220374, 'f1': 0.5093596059113301, 'number': 962}, 'overall_precision': 0.4377880184331797, 'overall_recall': 0.49536500579374276, 'overall_f1': 0.4648002174503941, 'overall_accuracy': 0.6648153886492026}
			------------EPOCH 8---------------
Loss:  tensor(566.4685, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(837.4218, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(697.4871, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(516.9731, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(608.0774, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(321.1670, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(341.8124, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(745.5363, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(594.5951, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(806.0490, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(813.2442, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(728.4331, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(459.1196, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(279.3556, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(207.5458, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(271.7220, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(315.4233, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(254.4673, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(220.5830, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(620.7847, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(666.4194, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(358.4448, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(419.1670, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(731.0026, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(731.8783, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(840.7535, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(288.6052, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(423.0515, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(459.3343, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(412.2119, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(549.9255, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(947.2382, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(735.3064, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(652.7563, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(350.6067, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(793.7982, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(503.6418, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.33636363636363636, 'recall': 0.43586387434554974, 'f1': 0.37970353477765106, 'number': 764}, 'P': {'precision': 0.36959761549925485, 'recall': 0.2577962577962578, 'f1': 0.3037354562155542, 'number': 962}, 'overall_precision': 0.349789283564118, 'overall_recall': 0.3366164542294322, 'overall_f1': 0.34307646885149096, 'overall_accuracy': 0.6083712726145926}
			------------EPOCH 9---------------
Loss:  tensor(449.7814, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(848.7159, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(642.7013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(529.5259, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(503.6543, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(299.4403, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(372.1917, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(735.2344, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(503.8179, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(730.8205, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(789.6398, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(724.4738, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(397.9560, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(180.1541, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(172.2844, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(245.9267, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(234.9073, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(187.7687, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(172.3247, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(499.7011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(593.8685, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(338.4332, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(374.2744, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(713.6577, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(641.2865, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(747.2139, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(321.2714, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(426.7901, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(390.2951, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(438.3966, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(576.1371, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(915.3775, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(813.7471, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(674.3890, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(320.1926, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(780.0889, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(474.4928, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4524207011686144, 'recall': 0.35471204188481675, 'f1': 0.3976522377109318, 'number': 764}, 'P': {'precision': 0.4508710801393728, 'recall': 0.6725571725571725, 'f1': 0.5398414685022945, 'number': 962}, 'overall_precision': 0.45132743362831856, 'overall_recall': 0.5318655851680185, 'overall_f1': 0.4882978723404255, 'overall_accuracy': 0.6891759826472481}
			------------EPOCH 10---------------
Loss:  tensor(366.7773, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(663.3032, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(484.7292, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(290.3743, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(331.5215, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(206.9346, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(199.4037, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(477.9709, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(337.2647, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(501.6974, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(556.0262, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(503.5834, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(354.6667, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(139.0393, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(175.7915, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(193.2187, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(238.8088, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(187.9889, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(196.6394, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(517.8107, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(660.9951, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(209.1064, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(449.5607, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(718.3483, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(541.2220, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(607.9831, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(190.9973, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(417.4458, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(326.9003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(284.8897, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(352.2971, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(506.9904, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(437.9511, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(411.4738, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(231.3599, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(537.9492, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(362.9555, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5138121546961326, 'recall': 0.12172774869109948, 'f1': 0.19682539682539685, 'number': 764}, 'P': {'precision': 0.4339063426200356, 'recall': 0.760914760914761, 'f1': 0.5526613816534541, 'number': 962}, 'overall_precision': 0.4416488222698073, 'overall_recall': 0.4779837775202781, 'overall_f1': 0.45909849749582643, 'overall_accuracy': 0.6506090148499512}
			------------EPOCH 11---------------
Loss:  tensor(664.9868, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(848.4407, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(839.9546, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(558.8986, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(330.7445, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(196.7426, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(207.3810, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(547.9628, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(295.9971, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(418.4330, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(497.7693, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(395.0647, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(228.2018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(108.7461, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(114.7734, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(149.1192, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(168.2580, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(132.9066, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(124.2742, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(359.0276, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(525.1982, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(235.1766, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(303.0622, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(505.8933, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(515.4220, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(787.6844, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(270.4730, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(508.6184, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(566.3864, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(567.1976, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(681.2592, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(747.0524, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(518.7324, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(409.1410, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(235.8848, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(578.9056, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(222.9469, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.38857142857142857, 'recall': 0.17801047120418848, 'f1': 0.24416517055655299, 'number': 764}, 'P': {'precision': 0.3719298245614035, 'recall': 0.5509355509355509, 'f1': 0.44407205697528274, 'number': 962}, 'overall_precision': 0.3752112676056338, 'overall_recall': 0.3858632676709154, 'overall_f1': 0.3804627249357327, 'overall_accuracy': 0.6525159106619312}
			------------EPOCH 12---------------
Loss:  tensor(273.6349, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(553.6063, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(352.6710, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(239.7539, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(387.6582, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(238.0843, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(221.7537, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(458.3347, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(405.3977, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(628.1507, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(751.2228, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(634.9028, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(320.8598, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(180.8459, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(175.5315, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(152.5246, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(358.3565, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(219.8121, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(221.4140, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(703.6140, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(620.0435, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(249.3022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(344.0405, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(516.9988, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(466.8219, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(488.1329, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(181.9340, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(226.7950, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(275.1213, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(236.0491, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(286.7237, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(424.5657, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(383.3305, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(292.8247, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(188.7115, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(546.4357, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(397.3583, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.31865828092243187, 'recall': 0.5968586387434555, 'f1': 0.4154897494305239, 'number': 764}, 'P': {'precision': 0.5877061469265368, 'recall': 0.4074844074844075, 'f1': 0.48127685696746475, 'number': 962}, 'overall_precision': 0.4041944709246902, 'overall_recall': 0.49130938586326767, 'overall_f1': 0.4435146443514645, 'overall_accuracy': 0.6356875551212071}
			------------EPOCH 13---------------
Loss:  tensor(291.5869, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(999.8447, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(459.3783, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(317.1500, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(329.4235, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(205.6124, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(335.1263, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(707.8293, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(310.2778, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(474.3468, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(635.8852, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(510.0043, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(178.7349, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(67.2407, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(101.3982, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(143.3541, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(119.8378, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(87.7692, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(79.3288, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(202.6046, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(273.1278, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(101.5199, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(153.8653, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(230.5568, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(256.5989, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(287.5650, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(128.8878, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(196.6820, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(195.1162, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(143.4075, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(224.4918, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(392.3663, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(285.8463, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(260.8652, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(140.0051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(383.9423, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(126.7013, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.504, 'recall': 0.3298429319371728, 'f1': 0.3987341772151899, 'number': 764}, 'P': {'precision': 0.4988847583643123, 'recall': 0.6975051975051975, 'f1': 0.5817078456870394, 'number': 962}, 'overall_precision': 0.5002710027100271, 'overall_recall': 0.5347624565469293, 'overall_f1': 0.5169420330439654, 'overall_accuracy': 0.6994732200319405}
			------------EPOCH 14---------------
Loss:  tensor(182.9246, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(286.9850, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(174.1797, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(85.7024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(92.2448, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(70.3940, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(78.4594, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(223.4831, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(143.3496, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(222.5901, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(271.4494, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(195.6650, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(122.5134, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(67.3439, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(76.8089, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(71.8740, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(122.9634, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(74.6024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(78.8760, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(168.5477, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(192.1241, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(79.7551, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(106.6197, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(166.4409, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(137.0227, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(170.1522, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(86.6727, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(117.9420, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(134.5772, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(91.4401, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(136.1372, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(153.1378, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(157.8369, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(139.7354, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(82.3870, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(251.4141, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(77.8561, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.44040862656072643, 'recall': 0.5078534031413613, 'f1': 0.47173252279635264, 'number': 764}, 'P': {'precision': 0.5593908629441624, 'recall': 0.5727650727650727, 'f1': 0.5659989727786339, 'number': 962}, 'overall_precision': 0.5032154340836013, 'overall_recall': 0.5440324449594438, 'overall_f1': 0.5228285077951002, 'overall_accuracy': 0.7037398994112459}
			------------EPOCH 15---------------
Loss:  tensor(141.6084, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(185.7322, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(139.1476, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.6387, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.8473, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.8067, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.5922, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(179.1162, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(114.5646, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(172.6246, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(244.1672, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(157.2510, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(77.6588, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.4736, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.3004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.4291, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(76.7657, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.7838, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.8794, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(105.7098, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(143.8737, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.1914, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(73.0082, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(139.5902, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(87.9760, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(120.3046, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.1013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(94.1281, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(77.4927, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(59.3304, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(109.1871, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(97.5610, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(89.3100, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(95.0045, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(59.5600, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(157.3680, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.3475, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.502906976744186, 'recall': 0.45287958115183247, 'f1': 0.4765840220385675, 'number': 764}, 'P': {'precision': 0.5469146238377007, 'recall': 0.6725571725571725, 'f1': 0.6032634032634033, 'number': 962}, 'overall_precision': 0.5307322287546766, 'overall_recall': 0.5753186558516802, 'overall_f1': 0.5521267723102585, 'overall_accuracy': 0.7096036040330846}
			------------EPOCH 16---------------
Loss:  tensor(108.4082, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(111.9100, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(86.4310, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.4481, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.2532, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.5425, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.1938, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(100.6377, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(98.9648, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(129.0557, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(194.3035, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(116.6761, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.8653, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.2399, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.5015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.0540, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(70.5074, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.7295, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.9097, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(86.3007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(110.1912, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.2113, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.5869, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(83.5049, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(71.0860, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(102.6432, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.5425, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(62.4405, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.9710, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.2688, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(89.1918, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(73.3662, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(64.1990, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(73.7231, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.2449, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(119.2677, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.8847, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4690721649484536, 'recall': 0.47643979057591623, 'f1': 0.4727272727272727, 'number': 764}, 'P': {'precision': 0.5381715362865221, 'recall': 0.5935550935550935, 'f1': 0.5645081562036578, 'number': 962}, 'overall_precision': 0.5089820359281437, 'overall_recall': 0.5417149478563151, 'overall_f1': 0.5248386191411731, 'overall_accuracy': 0.7011894262627226}
			------------EPOCH 17---------------
Loss:  tensor(83.6753, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(82.1945, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(69.2978, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.6262, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.0605, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.5575, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.9352, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(71.5925, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(87.1926, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(96.6758, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(135.1996, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(99.3365, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.0128, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.9452, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.8939, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.2896, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(62.9149, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.7491, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.6827, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.4705, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(89.5606, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.9478, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.5811, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(70.4836, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.1121, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(86.3920, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.6915, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.9961, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.1322, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.8806, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(73.4524, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.1714, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.9889, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(59.4476, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.3549, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(91.6909, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.1691, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4861294583883752, 'recall': 0.4816753926701571, 'f1': 0.4838921761998686, 'number': 764}, 'P': {'precision': 0.5587703435804702, 'recall': 0.6424116424116424, 'f1': 0.5976789168278531, 'number': 962}, 'overall_precision': 0.5292538915727322, 'overall_recall': 0.5712630359212051, 'overall_f1': 0.5494566731680133, 'overall_accuracy': 0.706028174385622}
			------------EPOCH 18---------------
Loss:  tensor(73.9115, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(67.2681, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(62.2582, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.9532, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.5507, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.8877, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.2543, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.2667, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(77.6010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(84.2604, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(119.9096, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(87.4766, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.0105, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.3543, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.6295, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.9705, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.2224, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.0320, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.7117, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.2292, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(77.4654, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.7137, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.3381, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(62.3329, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.2092, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(79.1653, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.4631, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.0294, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.8860, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.3496, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(64.8675, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.1307, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.9793, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.0782, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.1287, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(79.6993, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.8569, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.48226018396846254, 'recall': 0.4803664921465969, 'f1': 0.48131147540983604, 'number': 764}, 'P': {'precision': 0.5474654377880185, 'recall': 0.6174636174636174, 'f1': 0.5803615046409379, 'number': 962}, 'overall_precision': 0.5205850487540629, 'overall_recall': 0.5567786790266512, 'overall_f1': 0.5380739081746921, 'overall_accuracy': 0.7016184778204181}
			------------EPOCH 19---------------
Loss:  tensor(55.6384, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.4935, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.1267, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.2777, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.9985, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.3564, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.3744, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.5298, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(67.4047, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(73.7471, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(109.1040, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(72.2099, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.5924, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.7500, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.3454, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.6903, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.7419, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.2185, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.8524, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.1181, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(69.1876, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.6417, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.1627, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.4865, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.0860, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(73.6104, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.2214, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.2636, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.7236, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.6207, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.5301, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.4929, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.8203, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.7604, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.9604, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(72.2263, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.9415, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.48556430446194226, 'recall': 0.48429319371727747, 'f1': 0.4849279161205767, 'number': 764}, 'P': {'precision': 0.5505925250683683, 'recall': 0.6278586278586279, 'f1': 0.5866925692083536, 'number': 962}, 'overall_precision': 0.5239376008606778, 'overall_recall': 0.5643105446118193, 'overall_f1': 0.5433751743375174, 'overall_accuracy': 0.7035730460276977}
			------------EPOCH 20---------------
Loss:  tensor(51.5236, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.7500, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.8165, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.8455, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.0915, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.4825, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.4815, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.4918, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(59.4945, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(67.7641, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(103.2548, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.0778, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.5006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.4644, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.8008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.6493, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.6880, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.3604, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.1185, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.0086, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.8655, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.0525, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.6909, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.5679, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.8905, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(70.5444, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.0273, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.4353, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.3130, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.8873, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.1894, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.2716, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.7499, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.0840, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.6788, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(66.6676, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.0589, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.48429319371727747, 'recall': 0.48429319371727747, 'f1': 0.48429319371727747, 'number': 764}, 'P': {'precision': 0.5475319926873857, 'recall': 0.6226611226611226, 'f1': 0.5826848249027237, 'number': 962}, 'overall_precision': 0.5215285252960172, 'overall_recall': 0.5614136732329085, 'overall_f1': 0.5407366071428571, 'overall_accuracy': 0.7035015374347484}
			------------EPOCH 21---------------
Loss:  tensor(46.9415, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.1102, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.4716, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.9110, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.5147, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.9997, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.2520, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.6407, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.3201, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(62.5476, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(103.5672, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.8282, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.1400, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.0772, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.4501, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.1563, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.6018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.9675, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.6446, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.8281, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.5977, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.9632, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.6366, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.2850, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.8421, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(67.4817, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.5311, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.0296, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.0105, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.1730, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.8827, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.9090, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.2288, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.7731, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.7039, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.7074, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.7616, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.49001331557922767, 'recall': 0.4816753926701571, 'f1': 0.48580858085808576, 'number': 764}, 'P': {'precision': 0.5519713261648745, 'recall': 0.6403326403326404, 'f1': 0.5928777670837344, 'number': 962}, 'overall_precision': 0.527048741296197, 'overall_recall': 0.5701042873696408, 'overall_f1': 0.547731700528806, 'overall_accuracy': 0.7056706314208757}
			------------EPOCH 22---------------
Loss:  tensor(45.3139, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.9378, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.2198, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.9521, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.8014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.9677, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.1492, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.6592, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.5391, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.8861, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(115.2181, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.5993, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.1345, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.1367, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.3720, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.9988, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.2923, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.8081, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.6206, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.7857, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.9584, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.0986, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.3264, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.6481, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.4481, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(65.8554, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.4101, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.9101, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.8681, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.8559, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.5417, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.5913, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.4910, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.3406, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.0296, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.6623, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.0176, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4890038809831824, 'recall': 0.49476439790575916, 'f1': 0.49186727391021473, 'number': 764}, 'P': {'precision': 0.5525830258302583, 'recall': 0.6226611226611226, 'f1': 0.5855327468230694, 'number': 962}, 'overall_precision': 0.526117393645665, 'overall_recall': 0.5660486674391657, 'overall_f1': 0.5453530560982416, 'overall_accuracy': 0.7042404595618907}
			------------EPOCH 23---------------
Loss:  tensor(43.9176, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.2929, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.0489, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.0067, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.6794, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.9456, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.2528, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.1755, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.3772, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.5801, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(94.9779, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.2354, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.1023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.5302, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.5381, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.3954, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.2647, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.7610, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.4887, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.6573, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.6764, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.4360, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.4737, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.5282, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.4544, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(64.5892, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.4875, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.8885, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.0394, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.8853, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.6497, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.9004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.1142, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.4328, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.7364, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.5189, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.0519, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4792207792207792, 'recall': 0.4829842931937173, 'f1': 0.4810951760104302, 'number': 764}, 'P': {'precision': 0.5527272727272727, 'recall': 0.632016632016632, 'f1': 0.5897187196896218, 'number': 962}, 'overall_precision': 0.5224598930481283, 'overall_recall': 0.5660486674391657, 'overall_f1': 0.5433815350389323, 'overall_accuracy': 0.7052654160608299}
			------------EPOCH 24---------------
Loss:  tensor(42.7333, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.7952, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.9118, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.1436, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.8740, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.1916, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.4163, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.9216, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.5026, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.7038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(91.2723, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.9405, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.3898, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.0063, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.7681, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.6730, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.7731, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.2050, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.8884, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.5877, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.7246, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.8669, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.7353, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.4037, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.2822, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.1288, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.7880, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.8163, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.2168, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.0132, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.0305, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.4758, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.5954, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.3935, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.8939, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.1427, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.0720, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4856396866840731, 'recall': 0.4869109947643979, 'f1': 0.48627450980392156, 'number': 764}, 'P': {'precision': 0.5497287522603979, 'recall': 0.632016632016632, 'f1': 0.5880077369439072, 'number': 962}, 'overall_precision': 0.5235042735042735, 'overall_recall': 0.5677867902665121, 'overall_f1': 0.5447470817120623, 'overall_accuracy': 0.7031439944700022}
			------------EPOCH 25---------------
Loss:  tensor(42.0812, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.8283, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.2598, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.6488, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.1655, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.4514, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.6193, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.8472, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.9753, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.1203, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(87.9076, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.9091, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.8371, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.5806, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.9774, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.1654, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.4546, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.7605, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.5697, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.6967, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.5141, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.4800, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.1280, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.6999, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.7466, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(62.1257, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.2332, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.3973, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.4307, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.2069, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.5721, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.5006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.7496, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.6407, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.9278, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.9941, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.4139, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.49078947368421055, 'recall': 0.4882198952879581, 'f1': 0.48950131233595806, 'number': 764}, 'P': {'precision': 0.5493716337522442, 'recall': 0.6361746361746362, 'f1': 0.5895953757225434, 'number': 962}, 'overall_precision': 0.5256136606189968, 'overall_recall': 0.570683661645423, 'overall_f1': 0.5472222222222222, 'overall_accuracy': 0.7041212785736419}
			------------EPOCH 26---------------
Loss:  tensor(41.5702, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.1426, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.7384, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.1474, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.5226, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.8791, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.9513, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.0244, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.7841, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.9858, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(85.3968, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.7329, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.0056, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.3061, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.3484, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.7386, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.7601, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.8715, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.6993, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.9009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.2046, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.1319, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.6037, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.2285, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.2771, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(61.1790, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.8062, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.1285, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.1727, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.6677, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.6542, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.7297, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.3464, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.3672, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.3539, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.8961, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.7721, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4804177545691906, 'recall': 0.4816753926701571, 'f1': 0.48104575163398694, 'number': 764}, 'P': {'precision': 0.5441441441441441, 'recall': 0.6278586278586279, 'f1': 0.5830115830115831, 'number': 962}, 'overall_precision': 0.5181236673773987, 'overall_recall': 0.5631517960602549, 'overall_f1': 0.5397001665741255, 'overall_accuracy': 0.7032155030629514}
			------------EPOCH 27---------------
Loss:  tensor(40.7083, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.2305, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.0020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.6022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.0738, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.3166, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.3296, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.2203, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.8313, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.9701, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(83.0312, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.2719, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.4402, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.1515, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.0667, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.4902, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.8788, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.8203, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.4769, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.7790, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.5384, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.6133, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.2211, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.6872, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.2549, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.0825, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.2840, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.3255, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.4405, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.0388, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.6228, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.0568, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.7008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.5549, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.5375, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.1775, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.4785, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4847277556440903, 'recall': 0.47774869109947643, 'f1': 0.48121292023731055, 'number': 764}, 'P': {'precision': 0.5397526501766784, 'recall': 0.6351351351351351, 'f1': 0.5835721107927412, 'number': 962}, 'overall_precision': 0.5177718832891247, 'overall_recall': 0.5654692931633836, 'overall_f1': 0.5405704790916644, 'overall_accuracy': 0.7055037780373274}
			------------EPOCH 28---------------
Loss:  tensor(40.1961, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.9030, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.5151, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.1577, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.8238, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.8187, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.8363, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.7179, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.0957, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.1748, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(82.2735, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.3390, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.8745, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.7368, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.8852, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.0869, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.5478, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.9982, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.7439, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.8534, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.4246, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.5780, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.2161, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.2519, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.6616, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(59.5878, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.0430, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.8140, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.5389, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.6093, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.6123, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.6014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.7401, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.3882, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.1992, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.9227, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.1327, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4816272965879265, 'recall': 0.4803664921465969, 'f1': 0.48099606815203155, 'number': 764}, 'P': {'precision': 0.5424486148346738, 'recall': 0.6309771309771309, 'f1': 0.5833733781835656, 'number': 962}, 'overall_precision': 0.5178096757044125, 'overall_recall': 0.5643105446118193, 'overall_f1': 0.5400609925145551, 'overall_accuracy': 0.7035492098300479}
			------------EPOCH 29---------------
Loss:  tensor(39.5453, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.0398, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.9193, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.6490, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.5359, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.4167, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.4534, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.3847, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.7109, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.9891, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(82.4878, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.3958, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.9289, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.8547, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.1835, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.0671, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.4950, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.3198, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.2299, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.5022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.1982, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.1217, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.3248, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.8886, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.1718, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.8003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.7815, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.5239, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.0833, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.0542, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.4873, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.1848, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.0910, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.9490, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.6762, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.7960, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.8301, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5013477088948787, 'recall': 0.4869109947643979, 'f1': 0.4940239043824701, 'number': 764}, 'P': {'precision': 0.5527950310559007, 'recall': 0.6476091476091476, 'f1': 0.5964576352321684, 'number': 962}, 'overall_precision': 0.5323702514713751, 'overall_recall': 0.5764774044032445, 'overall_f1': 0.5535465924895688, 'overall_accuracy': 0.7067432603151146}
			------------EPOCH 30---------------
Loss:  tensor(41.7813, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.3156, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.6240, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.6810, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.0134, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.1110, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.9471, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.7756, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.7737, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.7950, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(81.8049, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.9992, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.3707, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.4490, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.5209, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.7502, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.6914, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.7650, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.6793, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.9428, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.8954, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.9102, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.9229, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.0337, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.5475, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(59.0170, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.7349, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.4111, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.0644, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.3860, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.4256, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.6290, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.0409, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.8861, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.8194, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.2622, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.7652, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.49024707412223667, 'recall': 0.49345549738219896, 'f1': 0.4918460534898891, 'number': 764}, 'P': {'precision': 0.5508166969147006, 'recall': 0.6309771309771309, 'f1': 0.5881782945736435, 'number': 962}, 'overall_precision': 0.5259219668626403, 'overall_recall': 0.5701042873696408, 'overall_f1': 0.5471226021684736, 'overall_accuracy': 0.705456105642028}
