Tokenizer: ../home/arg_mining/4epoch_complete/tokenizer Model: ../home/arg_mining/4epoch_complete/model
	Train size: 80 Test size: 20


		-------------RUN 1-----------
			------------EPOCH 1---------------
Loss:  tensor(3326.6809, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1950.9592, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2340.6860, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2247.0627, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1619.0464, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1586.6206, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1343.1256, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1684.3208, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2442.0615, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2291.0442, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1356.6848, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1206.7493, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1497.5308, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1927.6846, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2177.2786, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2700.8811, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2008.4178, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1515.1521, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2156.7307, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1376.7983, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1706.5669, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1479.1519, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1609.6075, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1124.9390, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2487.5500, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2364.9453, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2892.7026, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1417.4423, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2014.7209, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2727.2920, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2655.0903, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1334.4973, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1685.8209, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1757.3529, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2522.9822, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(935.2363, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(654.4258, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1807.9490, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1693.2324, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1216.6033, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1056.3878, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1541.8538, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1274.8230, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1602.7825, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1160.6631, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1752.0479, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2582.3518, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1397.6454, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1926.7356, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1577.8806, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1269.7474, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(968.7421, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1383.3228, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1323.3960, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2030.3743, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2034.0012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1201.8765, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.14583333333333334, 'recall': 0.075, 'f1': 0.09905660377358493, 'number': 280}, 'P': {'precision': 0.164, 'recall': 0.25153374233128833, 'f1': 0.19854721549636803, 'number': 326}, 'overall_precision': 0.15993788819875776, 'overall_recall': 0.16996699669966997, 'overall_f1': 0.1648, 'overall_accuracy': 0.5523283725396063}
			------------EPOCH 2---------------
Loss:  tensor(2081.5752, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1185.8735, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1396.2997, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1538.3748, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1499.5632, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1393.4781, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1068.4785, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1422.4832, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1627.0945, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1736.6017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1074.5212, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(850.0302, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1143.5621, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1553.6226, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1654.3674, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2126.9329, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1578.5476, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1116.8771, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1692.6080, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1182.9590, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1513.7520, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1043.2136, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1526.6484, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(930.8645, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2108.5105, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1932.7147, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2480.2122, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1092.0298, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1746.5782, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2254.7100, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2059.2915, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1108.8448, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1436.6808, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1298.4355, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2167.5952, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(777.0814, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(550.3580, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1634.1904, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1411.9071, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1012.3568, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(943.0402, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1323.7457, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1045.3645, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1406.0549, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(968.8615, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1537.1143, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2116.1736, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1162.3855, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1413.9926, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1215.0917, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(960.4102, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(893.5233, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1122.8207, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1039.7549, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1685.6465, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1844.3618, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1033.9639, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.23706896551724138, 'recall': 0.19642857142857142, 'f1': 0.21484375, 'number': 280}, 'P': {'precision': 0.26458333333333334, 'recall': 0.3895705521472393, 'f1': 0.315136476426799, 'number': 326}, 'overall_precision': 0.2556179775280899, 'overall_recall': 0.30033003300330036, 'overall_f1': 0.2761760242792109, 'overall_accuracy': 0.5961353816610657}
			------------EPOCH 3---------------
Loss:  tensor(1716.5327, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(857.2914, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1188.8960, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1269.2920, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1232.3655, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1149.8918, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(880.6912, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1246.6578, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1374.0426, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1520.9395, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(938.1152, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(716.7709, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(968.8793, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1291.0391, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1421.6094, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1763.7994, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1244.2926, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(906.6564, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1417.1975, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(972.9030, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1272.0989, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(878.8845, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1133.1027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(748.6160, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1865.2949, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1665.0557, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1982.2009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(819.5726, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1289.0410, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1873.1278, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1798.6200, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(943.0644, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1128.3165, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1074.8044, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1689.1216, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(634.8176, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(491.2501, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1323.8289, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1183.8069, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(884.0656, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(831.6439, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(979.1185, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(761.2126, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1164.8597, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(806.7587, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1247.9030, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1783.8423, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(920.0101, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1170.1417, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1041.6929, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(807.8414, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(636.6978, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(827.4091, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(733.7614, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1247.9727, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1463.1121, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(865.7084, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.2952029520295203, 'recall': 0.2857142857142857, 'f1': 0.29038112522686027, 'number': 280}, 'P': {'precision': 0.3675889328063241, 'recall': 0.5705521472392638, 'f1': 0.44711538461538464, 'number': 326}, 'overall_precision': 0.34234234234234234, 'overall_recall': 0.4389438943894389, 'overall_f1': 0.3846710050614606, 'overall_accuracy': 0.6119779164666347}
			------------EPOCH 4---------------
Loss:  tensor(1339.9088, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(660.4816, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(847.1738, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1090.0387, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(807.9115, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(688.6747, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(627.4602, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(897.8737, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1025.2727, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1350.6108, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(863.2153, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(512.2626, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(840.6198, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1024.8521, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1256.5222, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1585.3564, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(879.9985, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(708.1773, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1054.0757, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(731.5106, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(835.1473, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(699.0679, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(880.4830, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(596.2370, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1508.8871, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1272.9247, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1595.3982, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(579.0430, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1053.4731, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1424.1864, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1226.6471, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(728.0079, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(787.0981, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(732.8442, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1146.3494, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(469.5919, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(343.9239, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1046.1415, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(894.8472, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(667.3420, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(706.5022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(627.0211, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(441.1526, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(887.9736, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(582.7911, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(858.2367, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1224.1543, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(680.5573, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(730.6205, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(653.9855, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(517.0945, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(420.3741, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(576.9948, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(529.6348, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(848.3798, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(993.2464, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(669.5972, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3660377358490566, 'recall': 0.3464285714285714, 'f1': 0.3559633027522935, 'number': 280}, 'P': {'precision': 0.43735763097949887, 'recall': 0.588957055214724, 'f1': 0.5019607843137255, 'number': 326}, 'overall_precision': 0.41051136363636365, 'overall_recall': 0.4768976897689769, 'overall_f1': 0.4412213740458016, 'overall_accuracy': 0.6775084013442151}
			------------EPOCH 5---------------
Loss:  tensor(967.3914, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(451.6051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(597.5386, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(749.0709, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(323.2826, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(281.1652, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(362.7635, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(398.8952, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(797.1997, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1044.5410, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(534.5752, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(342.2810, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(655.0994, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(700.5931, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(916.3500, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1165.8751, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(600.9299, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(450.4038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(851.9502, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(547.5106, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(487.7831, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(448.6278, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(483.4362, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(392.6421, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(766.2996, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(891.4474, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1071.3391, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(312.6465, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(886.5880, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(930.0618, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(843.5743, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(464.5192, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(592.4210, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(563.6556, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(804.4486, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(372.9032, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(285.2274, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(662.7642, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(610.4100, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(465.7456, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(463.8889, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(409.1559, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(282.2017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(581.4839, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(394.4365, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(623.7504, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(807.5793, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(515.0661, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(401.9726, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(356.0697, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(286.5917, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(327.5349, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(348.7431, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(324.7748, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(568.7479, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(615.5812, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(340.1550, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.43891402714932126, 'recall': 0.3464285714285714, 'f1': 0.38722554890219557, 'number': 280}, 'P': {'precision': 0.46875, 'recall': 0.598159509202454, 'f1': 0.5256064690026955, 'number': 326}, 'overall_precision': 0.45839874411302983, 'overall_recall': 0.48184818481848185, 'overall_f1': 0.4698310539018504, 'overall_accuracy': 0.6915506481036966}
			------------EPOCH 6---------------
Loss:  tensor(764.8608, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(321.6901, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(459.6158, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(503.5860, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(251.2690, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(243.7817, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(235.0650, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(282.2061, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(590.8073, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(821.3699, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(466.2682, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(297.5706, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(429.6768, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(491.2322, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(454.8347, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(648.6414, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(379.8160, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(278.4646, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(567.2043, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(368.3912, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(372.8037, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(365.1597, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(355.9107, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(357.7863, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(798.1101, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(655.5230, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(926.1898, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(231.0838, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(874.1937, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(816.3956, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(683.1411, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(373.6129, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(397.9648, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(389.4941, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(472.8447, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(240.5799, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(283.1870, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(674.8771, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(615.7961, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(646.7705, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(560.6027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(383.8415, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(241.5079, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(640.9126, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(264.2285, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(688.2786, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(636.3542, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(390.1004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(197.1312, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(209.1012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(190.1869, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(163.6413, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(223.0914, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(167.5987, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(294.1147, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(414.1489, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(301.0671, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.36909871244635195, 'recall': 0.30714285714285716, 'f1': 0.3352826510721248, 'number': 280}, 'P': {'precision': 0.4305799648506151, 'recall': 0.7515337423312883, 'f1': 0.547486033519553, 'number': 326}, 'overall_precision': 0.41271820448877805, 'overall_recall': 0.5462046204620462, 'overall_f1': 0.47017045454545453, 'overall_accuracy': 0.6412626020163226}
			------------EPOCH 7---------------
Loss:  tensor(589.1364, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(249.0788, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(424.1224, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(432.9385, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(236.3547, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(154.3753, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(150.2660, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(268.4965, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(545.3023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(602.3121, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(394.5059, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(175.5428, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(406.9940, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(467.5926, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(409.8009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(549.8359, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(346.7411, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(231.9899, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(465.2898, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(326.9064, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(287.7451, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(277.2305, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(292.7936, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(250.4821, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(390.4495, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(356.1927, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(531.2184, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(170.4470, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(583.7927, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(470.9392, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(484.0048, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(183.1322, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(308.9777, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(276.9023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(350.6453, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(237.0865, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(207.2757, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(813.7850, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(588.8661, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(487.0386, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(364.7648, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(321.9012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(216.1110, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(461.5079, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(152.5256, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(289.3083, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(258.5920, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(193.9445, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(211.4664, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(262.7489, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(229.8340, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(137.0343, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(418.9386, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(406.4125, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(510.2549, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(667.0280, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(347.2043, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.45307443365695793, 'recall': 0.5, 'f1': 0.4753820033955858, 'number': 280}, 'P': {'precision': 0.5777027027027027, 'recall': 0.5245398773006135, 'f1': 0.5498392282958199, 'number': 326}, 'overall_precision': 0.5140495867768595, 'overall_recall': 0.5132013201320133, 'overall_f1': 0.5136251032204789, 'overall_accuracy': 0.7007321171387422}
			------------EPOCH 8---------------
Loss:  tensor(555.4905, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(327.8918, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(277.6845, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(376.0039, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(139.0365, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(104.7976, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.4437, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(102.8950, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(421.1973, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(438.3557, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(250.3556, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(116.1063, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(462.6818, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(658.5347, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(485.6590, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(561.9094, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(330.4704, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(235.3650, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(404.4090, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(279.0365, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(193.3731, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(207.3831, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(199.7682, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(112.2678, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(376.8573, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(350.8578, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(442.4337, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(115.0538, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(601.1771, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(789.5521, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(733.2540, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(255.4549, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(540.3073, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(606.4546, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(783.6206, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(386.0647, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(145.2948, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(379.0860, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(411.1163, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(192.4416, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(209.0284, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(85.4932, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(78.4031, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(200.9731, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(326.4423, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(571.0514, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(663.4215, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(329.5363, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(244.4472, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(262.7944, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(354.5644, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(260.0216, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(328.0828, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(165.4007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(273.2459, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(377.6804, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(256.0114, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.42134831460674155, 'recall': 0.26785714285714285, 'f1': 0.32751091703056767, 'number': 280}, 'P': {'precision': 0.4547738693467337, 'recall': 0.5552147239263804, 'f1': 0.5, 'number': 326}, 'overall_precision': 0.4444444444444444, 'overall_recall': 0.42244224422442245, 'overall_f1': 0.43316412859560066, 'overall_accuracy': 0.6647263562169947}
			------------EPOCH 9---------------
Loss:  tensor(428.1552, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(149.5630, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(262.2088, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(284.8515, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(122.4418, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(124.8097, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(120.1628, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(88.9055, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(376.3670, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(356.4575, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(241.0428, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(87.7808, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(173.5962, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(263.0468, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(270.1435, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(395.8697, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(241.9187, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(171.4332, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(225.2267, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(94.5068, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(135.4349, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(126.3024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(149.9420, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(97.4670, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(208.3450, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(316.1885, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(378.6937, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(100.2077, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(480.9666, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(208.6503, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(242.1894, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(98.1158, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(272.7604, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(384.5085, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(332.8220, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(187.4415, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(95.0854, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(388.6621, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(330.0538, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(231.8460, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(252.9590, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(103.0626, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(81.8528, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(306.1069, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(128.6778, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(173.8723, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(209.9511, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(132.0656, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(86.3376, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(83.9395, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(65.5924, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.4141, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(80.2232, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(70.8403, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(139.5707, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(109.3934, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(95.4097, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4280442804428044, 'recall': 0.4142857142857143, 'f1': 0.42105263157894735, 'number': 280}, 'P': {'precision': 0.4647058823529412, 'recall': 0.48466257668711654, 'f1': 0.47447447447447444, 'number': 326}, 'overall_precision': 0.44844517184942717, 'overall_recall': 0.4521452145214521, 'overall_f1': 0.45028759244042726, 'overall_accuracy': 0.6870499279884782}
			------------EPOCH 10---------------
Loss:  tensor(272.2970, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.7971, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(117.6888, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(132.2350, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(83.1339, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(100.3135, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.4205, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(62.5033, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(124.4292, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(256.8940, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(196.4866, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(67.0438, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(128.2531, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(259.6434, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(132.0608, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(214.9711, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(216.8073, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(195.8618, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(211.6311, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(148.2077, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(168.3195, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(74.5914, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(92.8275, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(109.8001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(88.1404, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(139.2548, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(204.2407, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(76.3530, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(242.5413, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(72.7803, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(90.7877, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.7455, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(122.8018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(174.7969, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(160.4780, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(70.7284, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(64.6842, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(209.9725, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(172.0538, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(108.5939, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(126.3048, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.1302, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.1976, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(115.9493, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(99.2625, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(168.6485, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(134.1986, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(106.1668, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(107.4983, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(82.2184, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.1417, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(198.3791, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(101.0994, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(98.7923, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(216.1885, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(102.5527, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(117.3456, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4444444444444444, 'recall': 0.6428571428571429, 'f1': 0.5255474452554744, 'number': 280}, 'P': {'precision': 0.573208722741433, 'recall': 0.5644171779141104, 'f1': 0.5687789799072642, 'number': 326}, 'overall_precision': 0.5013774104683195, 'overall_recall': 0.6006600660066007, 'overall_f1': 0.5465465465465466, 'overall_accuracy': 0.6847095535285646}
			------------EPOCH 11---------------
Loss:  tensor(217.0069, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(88.7396, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(98.5485, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(97.6254, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(82.8960, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.3588, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.9607, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.8534, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(76.7898, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(149.2044, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(93.1817, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.2763, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(92.5178, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(90.6414, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(61.2680, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(139.0487, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(76.1627, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(62.9187, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(125.6660, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.7537, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(71.1988, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.3524, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(64.8901, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.5384, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(70.0424, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(96.7076, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(116.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.7326, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(101.8340, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(78.1603, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(80.7377, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.1982, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(143.7999, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(172.7941, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(143.2932, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(61.3557, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.3040, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(119.6652, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(140.3841, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(85.0990, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(124.7595, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.8053, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.9639, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(73.9757, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.5072, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(91.0480, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(81.5311, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.8297, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.0153, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.4832, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.6216, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.5225, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.8351, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.0459, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(71.6135, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.4185, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.6100, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5232067510548524, 'recall': 0.44285714285714284, 'f1': 0.4796905222437137, 'number': 280}, 'P': {'precision': 0.5382882882882883, 'recall': 0.7331288343558282, 'f1': 0.6207792207792208, 'number': 326}, 'overall_precision': 0.5330396475770925, 'overall_recall': 0.599009900990099, 'overall_f1': 0.5641025641025641, 'overall_accuracy': 0.7342174747959673}
			------------EPOCH 12---------------
Loss:  tensor(150.1165, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.0034, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(59.7120, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(78.3841, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.1807, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.3101, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.7842, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.9302, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.3064, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(75.5337, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.7533, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.8614, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.1218, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(68.1503, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.2001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.4610, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.9795, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.9485, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(77.0401, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.3084, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.6301, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.0694, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.7474, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.0055, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(66.2782, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(78.5893, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(101.3745, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.3728, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.6786, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(61.3141, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.9987, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.3624, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.4179, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(90.9406, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.6263, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.2060, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.0718, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(93.7187, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(101.4379, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.8799, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(71.7214, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.9049, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.6573, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.1884, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.7759, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.1487, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.2147, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.9309, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.0204, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.0359, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.6803, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.0454, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.9982, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.0525, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.1842, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.4933, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.9924, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5198412698412699, 'recall': 0.46785714285714286, 'f1': 0.49248120300751885, 'number': 280}, 'P': {'precision': 0.4897959183673469, 'recall': 0.6625766871165644, 'f1': 0.5632333767926988, 'number': 326}, 'overall_precision': 0.5007215007215007, 'overall_recall': 0.5726072607260726, 'overall_f1': 0.5342571208622017, 'overall_accuracy': 0.7293566970715314}
			------------EPOCH 13---------------
Loss:  tensor(117.8363, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.5729, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.8040, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.6759, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.2458, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.1039, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.4399, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.9507, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.6091, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.7016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.1050, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.4930, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.1853, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.5501, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.4120, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.5655, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.3033, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.6787, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.0512, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.7929, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.0295, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.5881, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.5147, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.6350, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.4096, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.9619, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.5440, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.0976, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.7675, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.5664, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.4591, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.2156, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.8601, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(72.9642, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.2491, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.8465, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.5541, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.0660, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(72.7229, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.3423, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.5470, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.6091, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.4845, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.8387, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.3353, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.1356, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.6927, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.4187, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.2613, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.9117, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.6158, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.0021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.8395, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.4755, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.1788, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.1396, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.4622, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4983922829581994, 'recall': 0.5535714285714286, 'f1': 0.5245346869712353, 'number': 280}, 'P': {'precision': 0.5163934426229508, 'recall': 0.5797546012269938, 'f1': 0.546242774566474, 'number': 326}, 'overall_precision': 0.5081240768094535, 'overall_recall': 0.5676567656765676, 'overall_f1': 0.5362431800467654, 'overall_accuracy': 0.7231156985117618}
			------------EPOCH 14---------------
Loss:  tensor(100.7046, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.8185, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.6463, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.1885, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.7457, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.8342, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.8375, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.9925, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.2874, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.3597, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.6131, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.8657, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.8953, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.5687, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.5014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.1185, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.5227, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.8190, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.5680, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.9011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.4067, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.7008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.3547, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.9809, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.9827, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.4725, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.6809, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.8577, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.5583, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.4138, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.7122, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.5919, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.0793, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.3966, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.3647, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.5283, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.9374, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.0802, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.7068, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.7171, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.8845, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.4657, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.2973, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.0016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.8555, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.4481, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.9481, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.3765, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.9995, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.4577, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.5065, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.0918, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.0994, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.6347, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.0256, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.0991, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5179856115107914, 'recall': 0.5142857142857142, 'f1': 0.5161290322580646, 'number': 280}, 'P': {'precision': 0.5158924205378973, 'recall': 0.647239263803681, 'f1': 0.5741496598639455, 'number': 326}, 'overall_precision': 0.5167394468704513, 'overall_recall': 0.5858085808580858, 'overall_f1': 0.5491105955143079, 'overall_accuracy': 0.7199351896303409}
			------------EPOCH 15---------------
Loss:  tensor(84.6109, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.9725, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.2070, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.8492, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.3794, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.6342, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.5032, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.5819, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.6604, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.5306, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.1921, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.5846, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.5963, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.9106, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.8242, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.9615, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.6445, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.5657, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.9386, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.9942, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.3687, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.5368, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.5640, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.6732, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.4273, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.8358, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.5405, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.7176, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.6863, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.2908, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.1813, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.6403, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.2440, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.9370, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.9341, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.7123, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.3091, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.3864, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.3792, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.7890, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.2302, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.1350, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.4961, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.9970, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.0430, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.7386, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.4417, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.7259, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.6171, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.7947, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.2807, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.5443, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.4345, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.9197, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.5821, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.2877, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.3866, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.47686832740213525, 'recall': 0.4785714285714286, 'f1': 0.47771836007130125, 'number': 280}, 'P': {'precision': 0.5307692307692308, 'recall': 0.6349693251533742, 'f1': 0.5782122905027933, 'number': 326}, 'overall_precision': 0.5081967213114754, 'overall_recall': 0.5627062706270627, 'overall_f1': 0.5340642129992168, 'overall_accuracy': 0.721675468074892}
			------------EPOCH 16---------------
Loss:  tensor(70.8137, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.8050, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.4267, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.1863, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.8398, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.5182, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.2188, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.3698, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.5201, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.1429, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.9619, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.7155, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.6581, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.1462, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.3377, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.8453, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.2111, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.8542, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.2793, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.5714, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.6718, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.4202, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.7935, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.0662, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.4428, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.1175, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.0818, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.9009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.3827, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.2988, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.2875, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.5044, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.6853, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.6331, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.8568, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.0790, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.9332, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.4606, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.6598, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.7959, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.7358, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.2702, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.5652, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.7905, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.5750, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.4310, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.9972, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.6609, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.6491, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.3154, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.3895, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.5954, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.1430, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.3104, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.5892, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.4885, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.7478, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.505338078291815, 'recall': 0.5071428571428571, 'f1': 0.5062388591800357, 'number': 280}, 'P': {'precision': 0.5253807106598984, 'recall': 0.6349693251533742, 'f1': 0.5750000000000001, 'number': 326}, 'overall_precision': 0.5170370370370371, 'overall_recall': 0.5759075907590759, 'overall_f1': 0.5448868071818891, 'overall_accuracy': 0.7234157465194431}
			------------EPOCH 17---------------
Loss:  tensor(58.1766, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.6120, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.5999, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.5351, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.1986, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.5011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.8974, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.2408, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.0141, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.4754, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.0864, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.8333, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.3874, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.4921, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.0000, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.8306, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.7891, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.2309, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.9977, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.2159, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.8989, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.4649, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.8022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.3879, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.4707, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.4107, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.0793, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.4359, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.0078, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.9714, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.9660, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.6261, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.7950, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.6498, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.7553, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.7144, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.7388, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.9864, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.8603, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.0229, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.3111, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.5075, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.5379, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.1453, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.8533, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.8090, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.5662, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.7778, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.1123, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.0999, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.5330, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.9979, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.9055, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.8026, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.4145, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.9632, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.0490, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.48928571428571427, 'recall': 0.48928571428571427, 'f1': 0.48928571428571427, 'number': 280}, 'P': {'precision': 0.5210918114143921, 'recall': 0.6441717791411042, 'f1': 0.5761316872427983, 'number': 326}, 'overall_precision': 0.5080527086383602, 'overall_recall': 0.5726072607260726, 'overall_f1': 0.5384018619084562, 'overall_accuracy': 0.7210753720595295}
			------------EPOCH 18---------------
Loss:  tensor(47.0240, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.4044, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.1991, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.9399, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.2270, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.0462, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.3305, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.3655, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.0221, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.5610, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.1310, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.1918, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.4955, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.1601, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.4562, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.5143, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.5653, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.0763, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.0209, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.3501, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.6936, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.6721, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.3073, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.1773, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.9820, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.8466, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.0590, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.3213, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.0916, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.2229, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.1525, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.4190, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.4498, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.0731, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.0753, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.7420, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.0280, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.0836, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.0016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.7665, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.8541, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.0057, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.8756, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.7632, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.0078, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.4800, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.1747, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.4472, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.0494, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.2039, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.1737, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.4472, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.6941, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.8918, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.2219, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.6346, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.9486, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4823943661971831, 'recall': 0.48928571428571427, 'f1': 0.48581560283687947, 'number': 280}, 'P': {'precision': 0.5158150851581509, 'recall': 0.6503067484662577, 'f1': 0.5753052917232022, 'number': 326}, 'overall_precision': 0.5021582733812949, 'overall_recall': 0.5759075907590759, 'overall_f1': 0.5365103766333589, 'overall_accuracy': 0.7173547767642823}
			------------EPOCH 19---------------
Loss:  tensor(35.3883, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.7748, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.2369, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.2599, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.4461, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.1045, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.7914, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.5723, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.1855, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.5680, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.8280, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.7419, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.1839, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.2124, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.5860, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.8050, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.6587, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.1130, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.5986, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.6358, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.7640, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.0512, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.2201, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.2180, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.7371, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.7078, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.0305, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.5743, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.8213, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.7226, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.5757, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.5895, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.8777, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.6521, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.6822, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.0280, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.3924, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.9637, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.1741, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.7506, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.7312, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5782, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.3175, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.4147, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.3861, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.0602, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.9983, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.5509, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.2756, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.4584, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.0739, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.9938, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.6937, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.9410, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.3171, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.2193, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.9748, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4722222222222222, 'recall': 0.4857142857142857, 'f1': 0.47887323943661975, 'number': 280}, 'P': {'precision': 0.5208845208845209, 'recall': 0.6503067484662577, 'f1': 0.5784447476125513, 'number': 326}, 'overall_precision': 0.5007194244604316, 'overall_recall': 0.5742574257425742, 'overall_f1': 0.5349730976172175, 'overall_accuracy': 0.7128540566490639}
			------------EPOCH 20---------------
Loss:  tensor(27.2013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.1552, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.5610, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.9048, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.9463, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.0888, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.7813, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.8893, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.3451, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.0075, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.8782, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.3587, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.3163, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.3924, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.1691, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.5193, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.8800, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.2238, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.4992, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.0616, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.9132, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.5176, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.3869, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.5811, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.7637, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.8422, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.4639, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.9542, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.7771, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.3975, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.0234, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.1891, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.3864, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.6192, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.5145, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.0289, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.7340, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.4350, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.9453, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.9175, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2673, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.8961, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.7570, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.8818, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.6282, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.0598, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.8702, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.6955, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.8375, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.2607, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5980, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.8025, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.2810, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.6332, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.7134, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.2318, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.46875, 'recall': 0.48214285714285715, 'f1': 0.4753521126760563, 'number': 280}, 'P': {'precision': 0.5170731707317073, 'recall': 0.6503067484662577, 'f1': 0.5760869565217392, 'number': 326}, 'overall_precision': 0.497134670487106, 'overall_recall': 0.5726072607260726, 'overall_f1': 0.5322085889570553, 'overall_accuracy': 0.7108137301968315}
			------------EPOCH 21---------------
Loss:  tensor(22.6455, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.6962, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.0630, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.3916, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.6206, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.2678, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.3220, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.2517, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.3770, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.8147, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.1324, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.0590, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.6920, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.7659, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.6457, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.5479, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.2599, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.5078, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.6117, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.5463, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.2876, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.9302, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.6718, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.0500, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.9195, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.1071, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.4687, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.4738, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.0663, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.2734, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.6390, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.5396, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.2353, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.8883, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.7489, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.1134, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.7311, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.3294, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.3845, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.2972, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.1040, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.0124, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.5462, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.2738, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.4529, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.3001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.2738, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.3505, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.2654, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.3398, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.6160, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.3041, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.0600, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.6671, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.0145, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.2037, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.6601, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4755244755244755, 'recall': 0.4857142857142857, 'f1': 0.480565371024735, 'number': 280}, 'P': {'precision': 0.5194174757281553, 'recall': 0.656441717791411, 'f1': 0.5799457994579946, 'number': 326}, 'overall_precision': 0.501432664756447, 'overall_recall': 0.5775577557755776, 'overall_f1': 0.5368098159509204, 'overall_accuracy': 0.7125540086413826}
			------------EPOCH 22---------------
Loss:  tensor(19.4695, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.3418, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.5816, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.0696, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.4708, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.6119, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.1627, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.4896, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.9412, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.9041, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.5430, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.7688, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.1881, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.2442, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.1161, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.7868, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.7323, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.8651, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.8818, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.1205, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.7944, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.5149, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.1624, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.6482, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.2513, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.4610, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.7411, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.0623, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.4869, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.2926, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.4182, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.1744, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.5357, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.7422, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.0375, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.7801, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.5796, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.3083, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.6427, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.7039, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.2630, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.8050, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.2582, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.4212, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.1063, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.5218, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.6052, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.9330, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.8959, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.9125, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.0951, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.0291, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.4361, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.2017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.5305, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.7118, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.1951, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4788732394366197, 'recall': 0.4857142857142857, 'f1': 0.48226950354609927, 'number': 280}, 'P': {'precision': 0.5256723716381418, 'recall': 0.6595092024539877, 'f1': 0.5850340136054423, 'number': 326}, 'overall_precision': 0.5064935064935064, 'overall_recall': 0.5792079207920792, 'overall_f1': 0.5404157043879908, 'overall_accuracy': 0.7135141622659625}
			------------EPOCH 23---------------
Loss:  tensor(17.2760, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.9970, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.1760, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.1516, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.4611, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.0704, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.8039, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.0940, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.5478, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.1181, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.0281, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5438, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.7913, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.8319, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.7095, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.1804, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.3056, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.3442, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.2832, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.7225, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.4102, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.1690, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.7463, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.3126, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.0086, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.6289, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.1587, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.7127, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.0181, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.4504, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.4387, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.8622, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.9062, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.8559, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.4143, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.4960, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.6662, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.5330, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.0652, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.1247, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.9951, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6308, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.9932, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.7876, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.8177, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.9596, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.0341, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.6072, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.5844, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.5471, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.6538, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.8009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.9487, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.7598, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.0970, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.2806, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.8166, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.48056537102473496, 'recall': 0.4857142857142857, 'f1': 0.48312611012433393, 'number': 280}, 'P': {'precision': 0.526829268292683, 'recall': 0.6625766871165644, 'f1': 0.5869565217391305, 'number': 326}, 'overall_precision': 0.5079365079365079, 'overall_recall': 0.5808580858085809, 'overall_f1': 0.5419553502694381, 'overall_accuracy': 0.7141742678828612}
			------------EPOCH 24---------------
Loss:  tensor(15.5865, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.6885, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.7695, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.5415, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.5762, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.6477, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5718, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.7708, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.2285, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.4219, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.6115, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.3309, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.4339, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.4783, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.2885, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.6067, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.9421, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.8425, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.7887, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.3534, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.0633, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.8690, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.4151, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.0403, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.0444, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.8744, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.6520, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.4171, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.6013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.7765, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.6720, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.5883, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.0404, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.1325, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.8331, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.2389, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.9763, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.9036, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.5869, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.5714, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.7456, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4825, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.7545, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.2474, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.5491, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.4761, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.5372, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.3481, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.3144, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.2182, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.2559, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6189, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.5551, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.3436, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.7306, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.8381, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5074, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.48056537102473496, 'recall': 0.4857142857142857, 'f1': 0.48312611012433393, 'number': 280}, 'P': {'precision': 0.5230024213075061, 'recall': 0.6625766871165644, 'f1': 0.584573748308525, 'number': 326}, 'overall_precision': 0.5057471264367817, 'overall_recall': 0.5808580858085809, 'overall_f1': 0.5407066052227343, 'overall_accuracy': 0.7147743638982237}
			------------EPOCH 25---------------
Loss:  tensor(14.2022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.4239, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.3893, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.0563, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.8395, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.3058, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3820, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.4919, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.9639, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.7783, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.2237, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.1417, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.1199, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.1888, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.9134, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.1299, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.6381, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.4104, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.3552, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.9952, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.7641, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.6003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.1498, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.8283, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.2809, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.3480, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.2076, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.1472, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.2361, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.2015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.0687, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.3567, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.4680, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.5370, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.3136, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.0134, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.4372, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.3689, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.1394, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.0607, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.5028, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3578, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5470, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.7666, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.2825, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.0625, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.0768, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.1277, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.0724, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.9100, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.8871, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4659, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.2172, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.9604, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.3938, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.2867, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2548, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.47703180212014135, 'recall': 0.48214285714285715, 'f1': 0.47957371225577267, 'number': 280}, 'P': {'precision': 0.5205811138014528, 'recall': 0.6595092024539877, 'f1': 0.5818673883626523, 'number': 326}, 'overall_precision': 0.5028735632183908, 'overall_recall': 0.5775577557755776, 'overall_f1': 0.5376344086021505, 'overall_accuracy': 0.7131541046567451}
			------------EPOCH 26---------------
Loss:  tensor(12.9889, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2157, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.9997, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.6072, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.2451, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.0305, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2270, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.2558, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.7127, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.2501, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.8860, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.9534, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.8254, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.9097, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.5216, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.7717, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.3759, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.9907, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.9256, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.6801, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.3589, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.9394, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6627, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.6988, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.8979, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.8758, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.8813, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.9103, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.6958, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.5806, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.1696, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.9656, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.0295, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.8669, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.8148, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.0040, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.8832, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.7321, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.5852, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.7406, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2499, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.3650, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.3233, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.0466, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.6699, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.6481, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.9288, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.8362, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.6063, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.5597, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3168, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.9049, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.6740, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.1349, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.7807, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.0374, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4734982332155477, 'recall': 0.4785714285714286, 'f1': 0.47602131438721135, 'number': 280}, 'P': {'precision': 0.5193236714975845, 'recall': 0.6595092024539877, 'f1': 0.581081081081081, 'number': 326}, 'overall_precision': 0.5007173601147776, 'overall_recall': 0.5759075907590759, 'overall_f1': 0.5356868764389869, 'overall_accuracy': 0.7126740278444551}
			------------EPOCH 27---------------
Loss:  tensor(11.9765, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.0085, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.6599, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.2100, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.7261, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.8008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0974, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.0178, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.4959, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.7100, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5545, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.8056, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.5755, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.6986, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.2084, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.4427, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.1581, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.6559, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.5613, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.3495, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2840, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.1503, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.7523, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5146, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.2256, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.7216, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.5498, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.6452, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.6367, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.2792, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.1756, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.9849, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.4734, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.5927, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.4069, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.6189, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.6474, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.4840, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.3609, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.1319, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.2263, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1555, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.1839, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.9166, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.8145, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.3306, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.2595, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.7628, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.6317, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.3356, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.2557, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.6447, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.3755, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.8824, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.3011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.8609, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4734982332155477, 'recall': 0.4785714285714286, 'f1': 0.47602131438721135, 'number': 280}, 'P': {'precision': 0.528395061728395, 'recall': 0.656441717791411, 'f1': 0.585499316005472, 'number': 326}, 'overall_precision': 0.5058139534883721, 'overall_recall': 0.5742574257425742, 'overall_f1': 0.5378670788253477, 'overall_accuracy': 0.709493518963034}
			------------EPOCH 28---------------
Loss:  tensor(11.0186, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.8425, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.3038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.8357, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.2892, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6187, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9909, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.8146, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2868, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.2678, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2871, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6511, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.3305, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.4816, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.8649, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.1668, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.9684, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.3068, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.2092, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.0585, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.0860, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.9633, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3828, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.8484, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.8836, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.3022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.4100, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.3917, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.9046, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.8253, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.8147, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.0468, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.2027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.9884, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.4380, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.3505, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.1170, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.0072, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.6926, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.7979, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0746, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.0110, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.5441, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.6272, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.9689, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.8963, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6080, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.4194, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.0534, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.9862, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0820, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.3868, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.1722, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.6967, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.8998, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.7025, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.47183098591549294, 'recall': 0.4785714285714286, 'f1': 0.47517730496453897, 'number': 280}, 'P': {'precision': 0.5206812652068127, 'recall': 0.656441717791411, 'f1': 0.5807327001356852, 'number': 326}, 'overall_precision': 0.5007194244604316, 'overall_recall': 0.5742574257425742, 'overall_f1': 0.5349730976172175, 'overall_accuracy': 0.7101536245799328}
			------------EPOCH 29---------------
Loss:  tensor(10.2216, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6615, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.0204, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.5198, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.8930, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4583, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.5838, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.1109, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.7872, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.0214, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5324, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.1325, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.3269, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.6004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.8863, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.8073, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.0323, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.9162, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.7397, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.9137, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.8001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4648, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2697, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5518, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.3948, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.0304, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2071, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.1852, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.6160, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5361, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.6496, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.6309, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.8778, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.5343, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2531, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.1051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.8200, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.6715, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.2906, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.4729, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0055, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.8459, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.1853, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.4160, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.6676, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.5708, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4789, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2452, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.8132, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.7299, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9986, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.1759, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.9245, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.5196, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5771, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.47202797202797203, 'recall': 0.48214285714285715, 'f1': 0.47703180212014135, 'number': 280}, 'P': {'precision': 0.5297029702970297, 'recall': 0.656441717791411, 'f1': 0.5863013698630136, 'number': 326}, 'overall_precision': 0.5057971014492754, 'overall_recall': 0.5759075907590759, 'overall_f1': 0.5385802469135803, 'overall_accuracy': 0.709973595775324}
			------------EPOCH 30---------------
Loss:  tensor(9.4028, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5311, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.7173, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.2020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5706, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3341, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8251, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.3835, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.9257, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.4099, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.8221, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4033, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.9247, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.1468, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.2945, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.6329, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6590, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.7320, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.6182, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.4759, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.7662, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6478, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3523, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1618, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.3048, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.2433, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.8597, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.9865, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.9914, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.3167, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2697, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.4949, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.2431, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5968, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.1846, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.0966, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.8828, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5265, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.3197, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.9099, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.1647, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9409, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6830, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.8882, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2650, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.3000, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.2566, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3483, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.0473, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5488, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.4999, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8926, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.9389, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.7866, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.3654, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.1615, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4555, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4767025089605735, 'recall': 0.475, 'f1': 0.4758497316636851, 'number': 280}, 'P': {'precision': 0.5228915662650603, 'recall': 0.6656441717791411, 'f1': 0.5856950067476384, 'number': 326}, 'overall_precision': 0.5043227665706052, 'overall_recall': 0.5775577557755776, 'overall_f1': 0.5384615384615384, 'overall_accuracy': 0.7111137782045127}


		-------------RUN 2-----------
			------------EPOCH 1---------------
Loss:  tensor(1596.2292, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2367.3953, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2350.1643, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2018.7666, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2710.7632, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3295.2251, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3715.5703, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2818.6348, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1848.5044, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1282.4645, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2560.6025, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1774.8857, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(978.9894, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1736.7411, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1854.4156, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1924.1732, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1371.1542, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(502.8312, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1290.7863, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1759.5859, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(993.2281, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1129.5325, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(794.1893, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1434.9696, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(955.6156, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1559.8856, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2070.7935, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1626.5670, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2459.1958, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1226.2010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1210.3389, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2466.9268, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1106.1553, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(953.9185, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(653.6800, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(951.3903, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1973.0289, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1866.8790, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1699.3647, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1186.0641, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1153.1072, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(802.8644, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1771.1229, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(970.3476, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2044.7697, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1645.6624, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1484.1201, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(995.1294, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1744.1176, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1234.5896, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2095.2964, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1801.5466, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1601.1017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1127.9124, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(913.6257, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1393.9792, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1036.0200, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1886.1375, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.10597826086956522, 'recall': 0.1598360655737705, 'f1': 0.12745098039215688, 'number': 244}, 'P': {'precision': 0.25, 'recall': 0.04568527918781726, 'f1': 0.07725321888412016, 'number': 394}, 'overall_precision': 0.12954545454545455, 'overall_recall': 0.08934169278996865, 'overall_f1': 0.10575139146567718, 'overall_accuracy': 0.47823015669149754}
			------------EPOCH 2---------------
Loss:  tensor(1301.9290, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1681.6877, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1409.1265, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1914.0015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1781.9619, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2366.8965, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2648.9690, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1920.1821, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1533.8086, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(825.5544, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1729.1885, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1216.8716, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(707.9567, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1274.7346, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1466.8062, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1409.4265, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1071.5632, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(408.0446, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1039.8873, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1414.2902, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(772.4679, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(908.1189, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(637.6647, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1093.7045, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(752.1848, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1218.4211, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1655.5247, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1150.3788, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1984.4948, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(958.4520, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(920.5920, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1887.2932, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(908.2866, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(780.6157, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(579.2052, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(778.5378, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1278.7520, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1428.9226, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1360.1079, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(927.7206, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(869.9089, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(621.8991, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1305.3287, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(733.5044, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1657.6571, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1246.5460, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1027.3047, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(719.2829, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1172.6322, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(888.9727, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1620.3862, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1490.3735, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1312.0699, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(913.6677, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(698.5786, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(973.7856, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(865.9923, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1351.1061, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.2585227272727273, 'recall': 0.3729508196721312, 'f1': 0.3053691275167786, 'number': 244}, 'P': {'precision': 0.47346938775510206, 'recall': 0.29441624365482233, 'f1': 0.3630672926447574, 'number': 394}, 'overall_precision': 0.34673366834170855, 'overall_recall': 0.32445141065830724, 'overall_f1': 0.33522267206477735, 'overall_accuracy': 0.6244541484716157}
			------------EPOCH 3---------------
Loss:  tensor(1113.5688, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1320.4592, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1091.6235, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1616.0127, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1450.6802, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1824.1853, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2273.7253, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1446.0415, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1338.7491, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(637.3326, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1285.3303, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(817.4001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(562.6332, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(991.4985, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1192.8496, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1075.1326, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(818.7787, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(384.6520, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(860.9921, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1145.2987, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(619.5651, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(701.9335, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(496.2043, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(829.1373, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(614.3619, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(906.1194, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1173.4050, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(856.5578, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1424.4773, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(797.2684, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(675.5038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1339.2035, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(649.2001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(578.3632, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(436.5319, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(640.2903, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(981.8522, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1176.6993, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1080.9039, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(687.3134, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(657.0692, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(529.8478, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(976.5683, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(503.7924, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1199.1335, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(820.4570, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(580.9938, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(474.4723, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(737.5642, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(531.8671, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1067.7673, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1086.7551, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(996.0963, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(653.8712, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(465.7131, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(676.0906, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(579.9187, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1036.7804, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4128787878787879, 'recall': 0.44672131147540983, 'f1': 0.4291338582677166, 'number': 244}, 'P': {'precision': 0.6560283687943262, 'recall': 0.46954314720812185, 'f1': 0.5473372781065089, 'number': 394}, 'overall_precision': 0.5384615384615384, 'overall_recall': 0.4608150470219436, 'overall_f1': 0.4966216216216216, 'overall_accuracy': 0.7019650655021834}
			------------EPOCH 4---------------
Loss:  tensor(821.2120, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(909.2459, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(800.8420, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(850.7087, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1027.9092, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1255.7939, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1917.5986, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1063.4403, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(862.4995, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(470.4393, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1052.8232, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(471.3588, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(444.7731, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(675.0173, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1013.5459, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(809.7582, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(634.2635, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(252.9220, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(684.1133, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(816.1648, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(491.8331, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(549.6592, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(386.7649, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(604.1978, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(438.0032, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(625.1927, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(917.2073, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(720.3550, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1000.5297, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(571.8282, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(498.2471, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1059.8005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(478.9438, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(401.0722, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(320.3520, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(454.9060, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(717.7900, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(839.3932, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(958.9165, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(472.1405, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(611.6326, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(434.3055, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(758.0936, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(327.9860, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(867.0102, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(525.2178, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(319.4050, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(421.3254, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(375.3758, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(309.3169, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(685.9841, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(689.2124, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(729.1884, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(527.5974, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(379.1301, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(527.3832, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(462.1775, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(779.9277, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3438395415472779, 'recall': 0.4918032786885246, 'f1': 0.40472175379426645, 'number': 244}, 'P': {'precision': 0.6510067114093959, 'recall': 0.49238578680203043, 'f1': 0.5606936416184971, 'number': 394}, 'overall_precision': 0.4853168469860896, 'overall_recall': 0.49216300940438873, 'overall_f1': 0.48871595330739304, 'overall_accuracy': 0.6934240945286412}
			------------EPOCH 5---------------
Loss:  tensor(610.0819, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(718.9551, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(560.9170, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(487.3815, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(788.2523, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(861.9530, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1275.3989, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(699.3318, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(582.8444, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(433.7979, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(778.0533, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(322.4915, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(289.8566, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(464.0370, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(734.1233, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(561.3621, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(385.6671, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(174.2939, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(491.5797, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(569.1272, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(336.0990, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(424.3315, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(278.5694, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(418.5567, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(278.7988, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(492.1843, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(593.5724, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(466.8422, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(680.7913, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(446.1154, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(340.1015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(617.2449, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(279.3145, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(209.5148, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(210.5770, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(243.1540, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(467.1329, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(531.0070, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(625.2069, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(305.3958, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(377.8596, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(244.8662, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(465.8848, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(224.8362, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(588.3074, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(305.9939, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(197.2668, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(327.4663, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(290.2520, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(189.5358, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(359.3293, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(535.3836, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(429.4684, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(333.0649, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(210.1979, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(293.8194, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(366.1037, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(648.9866, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3236009732360097, 'recall': 0.5450819672131147, 'f1': 0.4061068702290076, 'number': 244}, 'P': {'precision': 0.66, 'recall': 0.16751269035532995, 'f1': 0.2672064777327935, 'number': 394}, 'overall_precision': 0.38943248532289626, 'overall_recall': 0.31191222570532917, 'overall_f1': 0.34638816362053965, 'overall_accuracy': 0.5524017467248908}
			------------EPOCH 6---------------
Loss:  tensor(497.9671, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(816.8969, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(713.5872, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(360.2780, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(675.5087, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(540.9937, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1004.5764, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(384.2161, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(398.4249, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(267.0570, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(688.7733, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(206.4077, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(387.6422, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(514.3268, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(781.2142, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(581.1703, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(331.8158, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(145.4087, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(408.1720, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(495.5568, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(314.5521, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(327.2288, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(237.5525, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(350.1456, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(189.7890, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(348.4938, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(550.1628, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(425.9031, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(600.9796, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(380.3854, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(302.4685, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(672.5768, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(199.0983, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(156.8658, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(185.0936, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(181.1321, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(308.5584, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(305.7248, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(393.7985, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(218.4608, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(278.7677, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(164.3069, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(396.9030, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(186.1355, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(428.9535, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(241.7293, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(164.9211, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(206.9622, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(214.3817, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(147.6457, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(309.1926, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(286.1993, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(439.4229, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(321.0555, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(142.0432, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(308.9644, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(183.4027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(579.7780, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.35674157303370785, 'recall': 0.5204918032786885, 'f1': 0.42333333333333334, 'number': 244}, 'P': {'precision': 0.7285714285714285, 'recall': 0.3883248730964467, 'f1': 0.5066225165562913, 'number': 394}, 'overall_precision': 0.49469964664310956, 'overall_recall': 0.438871473354232, 'overall_f1': 0.46511627906976744, 'overall_accuracy': 0.6466735165681994}
			------------EPOCH 7---------------
Loss:  tensor(290.1390, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(357.3344, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(356.1296, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(120.2625, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(507.1517, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(389.9953, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(884.8657, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(407.3145, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(343.4611, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(393.9965, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(735.6207, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(282.3189, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(166.8205, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(313.1698, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(497.2356, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(356.7877, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(250.8210, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(102.9483, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(264.2077, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(295.9610, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(290.1886, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(316.0946, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(275.0104, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(356.2468, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(308.2464, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(541.5150, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(576.9208, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(465.7520, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(820.3431, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(598.0724, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(383.0877, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(771.7153, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(160.6204, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(140.1385, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(141.6111, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(161.9021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(294.4510, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(375.0212, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(383.2213, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(193.3947, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(294.5232, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(140.2179, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(354.9849, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(360.6139, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(847.6414, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(432.1708, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(197.2504, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(151.2018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(185.0115, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(205.2124, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(355.1625, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(452.3959, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(361.0285, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(232.6553, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(146.8830, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(203.9066, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(180.4532, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(326.6446, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.47058823529411764, 'recall': 0.36065573770491804, 'f1': 0.40835266821345706, 'number': 244}, 'P': {'precision': 0.6430517711171662, 'recall': 0.5989847715736041, 'f1': 0.6202365308804205, 'number': 394}, 'overall_precision': 0.5848375451263538, 'overall_recall': 0.5078369905956113, 'overall_f1': 0.5436241610738254, 'overall_accuracy': 0.7277806319034164}
			------------EPOCH 8---------------
Loss:  tensor(327.0860, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(303.3428, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(336.7787, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(186.8341, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(578.2055, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(502.0519, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1111.6709, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(357.1946, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(377.4737, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(253.1240, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(467.5208, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(99.0934, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(154.9133, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(175.2688, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(334.0781, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(218.2237, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(270.0709, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(158.9968, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(348.7418, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(374.9557, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(181.5589, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(284.9036, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(219.2182, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(322.6097, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(74.6437, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(302.4304, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(520.4653, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(370.0980, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(364.4962, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(169.0012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(169.9855, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(393.9806, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(180.2705, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(164.6081, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(138.2449, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(218.0264, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(449.7777, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(562.2278, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(773.8929, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(335.3515, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(456.8275, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(421.3412, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(642.8114, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(200.4460, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(433.0115, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(186.9146, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(159.3147, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(274.9341, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(161.8378, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(89.3281, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(175.0705, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(137.1259, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(184.4356, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(164.4536, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(70.1493, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(111.7640, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(131.2955, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(259.7978, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.375, 'recall': 0.5532786885245902, 'f1': 0.4470198675496689, 'number': 244}, 'P': {'precision': 0.6260162601626016, 'recall': 0.19543147208121828, 'f1': 0.29787234042553196, 'number': 394}, 'overall_precision': 0.4389233954451346, 'overall_recall': 0.3322884012539185, 'overall_f1': 0.3782337198929528, 'overall_accuracy': 0.5540714102234781}
			------------EPOCH 9---------------
Loss:  tensor(193.9001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(290.8715, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(276.0146, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(66.2490, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(472.1431, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(409.1208, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(779.1029, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(293.2898, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(257.4714, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(253.9704, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(272.1487, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(134.1495, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(129.2094, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(120.7419, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(253.2161, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(183.3609, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(179.6212, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(84.8984, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(168.8991, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(177.6118, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(155.0667, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(136.4241, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(89.8094, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(143.4018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.9943, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(197.3640, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(237.7323, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(191.7868, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(210.6062, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(117.0761, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(111.1026, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(185.1893, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(87.0233, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(90.0954, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(102.0789, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(76.2732, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(180.6337, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(146.3347, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(280.1894, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(129.7073, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(220.4174, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(116.5481, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(336.4534, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(95.6019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(418.7864, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(242.0502, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(186.0185, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(206.2850, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(448.0836, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(157.7680, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(324.7804, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(346.9205, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(467.3420, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(196.5166, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(93.9521, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(187.9070, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(94.9466, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(211.0889, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5, 'recall': 0.6188524590163934, 'f1': 0.5531135531135531, 'number': 244}, 'P': {'precision': 0.6792452830188679, 'recall': 0.5482233502538071, 'f1': 0.6067415730337079, 'number': 394}, 'overall_precision': 0.5919354838709677, 'overall_recall': 0.5752351097178683, 'overall_f1': 0.5834658187599364, 'overall_accuracy': 0.7394040585666581}
			------------EPOCH 10---------------
Loss:  tensor(116.4986, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(124.9340, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(97.3371, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.7678, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(125.0631, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(101.7334, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(309.2257, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(103.3061, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(170.0307, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(165.2349, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(189.9417, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(64.4348, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(130.5562, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(129.9810, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(381.5809, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(294.2223, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(240.8313, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(159.6532, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(354.2373, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(386.4724, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(180.2078, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(181.3923, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(158.7737, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(251.9744, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.9288, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(190.3972, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(239.0640, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(394.1214, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(179.9380, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(95.0892, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(145.2009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(200.9215, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(74.3252, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(62.3011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(76.3171, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.7091, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(124.9782, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(85.8340, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(182.6198, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(105.4845, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(167.8266, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(87.6163, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(197.6413, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.2736, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(222.7227, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(101.3843, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(112.8209, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(133.9906, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(350.7622, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(141.3929, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(184.9435, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(144.1394, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(301.8802, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(233.4444, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(137.0546, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(183.0528, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(255.0957, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(492.6530, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.49140893470790376, 'recall': 0.5860655737704918, 'f1': 0.5345794392523364, 'number': 244}, 'P': {'precision': 0.6627358490566038, 'recall': 0.7131979695431472, 'f1': 0.687041564792176, 'number': 394}, 'overall_precision': 0.593006993006993, 'overall_recall': 0.664576802507837, 'overall_f1': 0.6267553584626756, 'overall_accuracy': 0.7523760595941433}
			------------EPOCH 11---------------
Loss:  tensor(536.1871, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(454.7916, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(141.3178, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(556.9714, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(334.8532, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(230.9142, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(476.4277, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(227.7725, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(191.4000, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(100.0569, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(211.2721, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(72.9786, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(82.5746, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(84.6777, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(122.9299, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(102.1548, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(159.5694, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(77.5964, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(157.1328, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(148.9549, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(103.4723, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(144.3117, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(118.4984, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(164.6727, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.2343, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(211.5741, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(513.1047, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(374.9313, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(688.8613, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(318.1269, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(374.5043, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(538.7900, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(252.9317, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(196.0692, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(127.7914, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(299.0046, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(318.0533, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(187.3467, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(282.7779, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(219.5633, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(106.3470, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.8210, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(159.5035, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.7314, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(99.1779, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(62.9069, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(85.4412, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(78.7888, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(90.7865, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(74.9292, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(125.2149, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(84.7211, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(220.3634, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(192.2009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.0871, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(184.4832, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(100.0339, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(422.7520, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.6850393700787402, 'recall': 0.35655737704918034, 'f1': 0.4690026954177897, 'number': 244}, 'P': {'precision': 0.5500878734622144, 'recall': 0.7944162436548223, 'f1': 0.6500519210799585, 'number': 394}, 'overall_precision': 0.5747126436781609, 'overall_recall': 0.6269592476489029, 'overall_f1': 0.5997001499250374, 'overall_accuracy': 0.7620729514513229}
			------------EPOCH 12---------------
Loss:  tensor(316.3658, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(233.7622, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(159.6110, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(194.4816, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(231.5354, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(174.7557, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(468.2391, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(136.3327, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(220.9487, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(101.9386, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(213.8080, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(106.4706, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(102.3231, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(299.9281, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(225.9048, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(139.3207, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(244.6439, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(97.6845, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(184.1339, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(244.9894, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(220.0408, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(188.3883, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(166.3676, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(313.4668, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(119.2217, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(299.2990, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(435.3289, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(430.8420, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(379.7874, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(229.6310, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(250.9108, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(582.1346, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(88.1167, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(77.4707, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(141.3415, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(64.4443, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(123.9821, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(77.1793, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(147.0070, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(114.1717, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(115.3470, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(62.9519, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(236.2302, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(70.7487, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(178.8266, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(107.3683, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(91.9299, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(80.9719, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(126.0618, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(79.0117, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(129.9034, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(160.7106, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(268.4918, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(203.3491, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(94.1828, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(152.5502, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(68.7915, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(247.4247, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.2857142857142857, 'recall': 0.06557377049180328, 'f1': 0.10666666666666666, 'number': 244}, 'P': {'precision': 0.5841035120147874, 'recall': 0.8020304568527918, 'f1': 0.6759358288770054, 'number': 394}, 'overall_precision': 0.5561139028475712, 'overall_recall': 0.5203761755485894, 'overall_f1': 0.5376518218623482, 'overall_accuracy': 0.722964294888261}
			------------EPOCH 13---------------
Loss:  tensor(474.8564, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(249.1306, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(199.1661, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(161.4462, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(897.3148, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(748.8860, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1680.0388, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1121.6646, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(176.9199, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(109.8261, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(250.2045, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(77.5034, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.9504, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.1733, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(124.2030, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.3143, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(93.4015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.8780, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(88.4253, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(84.6787, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(103.6019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(67.4121, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(84.5677, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(158.8580, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(648.0804, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(430.5053, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1005.0072, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(481.9810, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1154.9067, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(355.4451, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(191.7737, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(579.0494, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(368.2949, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(187.2260, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(254.3043, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(274.9682, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(351.0729, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(339.5915, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(464.9020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(264.8618, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(167.7774, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(61.1285, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(197.8359, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(72.7833, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(93.4293, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.5927, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.1395, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.2504, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.8308, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.6094, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(91.8769, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.4533, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(131.0130, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(119.2122, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(66.2106, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(110.4523, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(93.8589, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(226.3286, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5102040816326531, 'recall': 0.10245901639344263, 'f1': 0.17064846416382254, 'number': 244}, 'P': {'precision': 0.5015822784810127, 'recall': 0.8045685279187818, 'f1': 0.6179337231968811, 'number': 394}, 'overall_precision': 0.5022026431718062, 'overall_recall': 0.5360501567398119, 'overall_f1': 0.5185746777862016, 'overall_accuracy': 0.7010017980991523}
			------------EPOCH 14---------------
Loss:  tensor(572.7269, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(416.6613, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(365.1494, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(546.9045, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(698.1302, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(635.1947, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1144.6201, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(611.9662, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(582.3860, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(354.9696, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(698.2409, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(189.1201, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(376.6974, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(418.1609, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(632.4269, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(515.1122, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(165.1377, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(71.4333, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(413.8073, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(227.5103, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(83.8873, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.5849, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(80.7161, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(79.1500, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.6990, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(120.2974, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(156.8689, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(184.3320, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(208.2744, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(78.9240, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(70.4982, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(116.3413, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(152.4733, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(102.3792, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(149.2646, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(81.6866, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(448.4836, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(342.3082, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(484.2783, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(361.2300, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(596.3445, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(141.9023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1026.6168, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1213.2056, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1238.1411, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(636.9702, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(228.4395, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(200.3869, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(127.4751, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(721.4702, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1019.3525, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(448.5932, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(378.8317, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(304.6232, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(142.4406, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(196.0182, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(262.8491, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(259.1083, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5379061371841155, 'recall': 0.610655737704918, 'f1': 0.5719769673704413, 'number': 244}, 'P': {'precision': 0.6807387862796834, 'recall': 0.6548223350253807, 'f1': 0.6675291073738681, 'number': 394}, 'overall_precision': 0.6204268292682927, 'overall_recall': 0.6379310344827587, 'overall_f1': 0.6290571870170016, 'overall_accuracy': 0.7873105574107372}
			------------EPOCH 15---------------
Loss:  tensor(163.3456, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(244.1470, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(275.3837, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(104.4134, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(200.5534, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(178.1230, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(299.1594, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(135.0165, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(140.8337, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(99.2311, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(188.3313, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.6387, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(217.3543, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(244.3566, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(317.3024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(220.7514, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(464.8015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(139.6317, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(570.3300, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(499.8737, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(367.1833, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(255.4908, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(188.8355, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(307.4904, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(105.9462, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(341.2870, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(460.0884, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(264.0654, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(361.1190, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(550.4977, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(207.0667, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(354.9347, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(92.2712, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(106.4995, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(95.0601, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(80.8078, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(113.3381, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(166.8657, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(336.0762, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(127.0568, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(122.1147, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(98.7937, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(182.9512, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(74.4252, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(139.4085, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(73.5638, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(73.2935, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.9063, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.2763, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(87.5418, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(133.6527, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(115.3171, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(163.6917, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(99.4276, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(78.0627, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(76.8384, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(144.3067, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(177.4635, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.43333333333333335, 'recall': 0.6926229508196722, 'f1': 0.5331230283911672, 'number': 244}, 'P': {'precision': 0.7510548523206751, 'recall': 0.4517766497461929, 'f1': 0.5641838351822505, 'number': 394}, 'overall_precision': 0.5534290271132376, 'overall_recall': 0.5438871473354232, 'overall_f1': 0.5486166007905139, 'overall_accuracy': 0.7104418186488569}
			------------EPOCH 16---------------
Loss:  tensor(158.3251, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(275.5137, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(281.7192, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.1229, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(478.2308, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(265.8807, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(378.3354, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(206.5274, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(129.8675, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(141.9113, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(172.2713, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(188.6464, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(82.4000, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(86.7285, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(146.6797, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(116.2165, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(95.1426, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.8019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(100.7085, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(93.3244, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(76.7233, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.0767, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.5280, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.9226, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.0316, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(96.9766, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(131.1263, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(104.4335, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(92.0219, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(72.8533, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.0251, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(80.5215, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.6830, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.1061, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.2100, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.3102, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.7414, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.1422, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(100.9858, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(62.1560, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.7854, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.8597, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(112.8920, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.7047, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.1660, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.3827, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.8350, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.6856, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.1309, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.4530, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.3749, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.2342, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(64.3741, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.8883, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.1068, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.9547, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.8347, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.6374, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.6291079812206573, 'recall': 0.5491803278688525, 'f1': 0.5864332603938731, 'number': 244}, 'P': {'precision': 0.6900726392251816, 'recall': 0.7233502538071066, 'f1': 0.7063197026022304, 'number': 394}, 'overall_precision': 0.6693290734824281, 'overall_recall': 0.6567398119122257, 'overall_f1': 0.6629746835443038, 'overall_accuracy': 0.7970716670947855}
			------------EPOCH 17---------------
Loss:  tensor(54.0898, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.2296, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.4522, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.0854, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(75.1602, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(62.1325, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(138.5544, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.9246, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.7115, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.3006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.9044, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.8590, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.2161, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.5969, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(68.7753, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.9192, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(76.0785, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.3983, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(83.2752, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(61.7049, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.6910, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.4644, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.1241, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.6142, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.7613, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(76.0512, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(93.0693, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(98.6601, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(62.2366, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.5626, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.2888, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(68.5429, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.1525, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.4448, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.1365, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.0056, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.4564, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.9019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(68.3021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.0503, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.7430, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.6519, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(61.2542, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.1094, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.6454, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.1841, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.4308, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.8001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.0306, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.4911, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.1287, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.0666, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.9366, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.4120, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.5423, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.7161, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.9823, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.1008, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.569672131147541, 'recall': 0.569672131147541, 'f1': 0.569672131147541, 'number': 244}, 'P': {'precision': 0.720626631853786, 'recall': 0.700507614213198, 'f1': 0.7104247104247103, 'number': 394}, 'overall_precision': 0.6618819776714514, 'overall_recall': 0.6504702194357367, 'overall_f1': 0.6561264822134388, 'overall_accuracy': 0.7996403801695351}
			------------EPOCH 18---------------
Loss:  tensor(40.1888, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.6748, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.0722, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.6720, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.8893, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.8639, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(97.3150, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.1113, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.8892, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.8790, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.8683, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.2712, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.2536, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.1293, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.9909, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.2367, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.7571, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.1411, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.2637, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.4642, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.2430, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.3475, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.8616, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.4504, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.9264, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.4140, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(75.0673, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(59.2832, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.7840, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.3942, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.4626, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.2144, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.8076, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.8875, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.9959, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.0081, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.2390, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.1560, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.3323, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.1417, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.7357, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.0523, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.9709, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.1780, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.7977, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.5584, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.6988, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.5854, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.7744, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.3763, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.7358, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.3803, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.6213, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.1387, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.6437, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.9724, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.6814, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.0123, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.571969696969697, 'recall': 0.6188524590163934, 'f1': 0.5944881889763779, 'number': 244}, 'P': {'precision': 0.7300275482093664, 'recall': 0.6725888324873096, 'f1': 0.7001321003963012, 'number': 394}, 'overall_precision': 0.6634768740031898, 'overall_recall': 0.6520376175548589, 'overall_f1': 0.6577075098814229, 'overall_accuracy': 0.7925122013871051}
			------------EPOCH 19---------------
Loss:  tensor(31.7635, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.9249, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.5002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.2918, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.4033, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.8038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(74.2832, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.4428, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.2296, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.4524, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.8393, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.5743, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.7479, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.7397, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.6832, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.3865, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.2812, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.0388, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.2925, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.8439, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.8812, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.9069, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.0252, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.1583, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.1688, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.0102, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.8477, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.0474, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.8950, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.6196, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.3247, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.5934, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.9342, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.5052, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.4157, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.3278, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.1932, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.1110, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.7259, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.2862, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.1839, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.9277, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.8449, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.5239, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.0060, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.3403, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.4896, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.1269, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.0134, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.9754, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.5169, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.0027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.7049, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.5993, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.6723, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.1886, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.9413, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.4424, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.573076923076923, 'recall': 0.610655737704918, 'f1': 0.5912698412698413, 'number': 244}, 'P': {'precision': 0.7315068493150685, 'recall': 0.6776649746192893, 'f1': 0.7035573122529644, 'number': 394}, 'overall_precision': 0.6656, 'overall_recall': 0.6520376175548589, 'overall_f1': 0.6587490102929533, 'overall_accuracy': 0.7936681222707423}
			------------EPOCH 20---------------
Loss:  tensor(25.6968, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.2923, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.5379, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.4959, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.4366, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.2071, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(62.6015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.8427, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.8410, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.0959, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.6471, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.4385, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.9176, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.3847, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.8545, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.9125, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.5236, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.7166, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.1356, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.2195, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.6422, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.1398, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.1052, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.1020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.1883, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.0554, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.8298, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.9012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.7555, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.0144, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.1818, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.0347, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.5504, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.0510, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.3760, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.5856, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.7932, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.7427, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.2453, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.2853, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.2742, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.3065, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.1032, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.9633, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.5437, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.3819, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.0789, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.3190, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.0168, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.2236, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.0749, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.1332, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.9701, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.1386, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.1868, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.6975, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.8706, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.3495, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5658914728682171, 'recall': 0.5983606557377049, 'f1': 0.5816733067729084, 'number': 244}, 'P': {'precision': 0.7295081967213115, 'recall': 0.6776649746192893, 'f1': 0.7026315789473684, 'number': 394}, 'overall_precision': 0.6618589743589743, 'overall_recall': 0.6473354231974922, 'overall_f1': 0.6545166402535657, 'overall_accuracy': 0.7930901618289237}
			------------EPOCH 21---------------
Loss:  tensor(22.6965, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.4901, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.9991, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.9321, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.8739, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.5380, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.4351, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.1763, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.5391, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.7890, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.0840, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.7531, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.3707, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.0419, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.7676, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.7290, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.1547, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.0237, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.7131, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.5074, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.2455, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.5017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.0510, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.1513, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.5566, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.6245, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.8141, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.5141, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.7183, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.6339, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.7568, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.6280, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.2716, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.7987, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.0667, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.4980, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.0320, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.9921, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.2654, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.7358, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.9187, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.0938, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.8432, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.4433, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.5774, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.8492, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.2581, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.7208, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.4095, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.8454, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.3203, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.3709, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.9497, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.8470, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.8264, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.0927, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.6232, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.4347, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5348837209302325, 'recall': 0.5655737704918032, 'f1': 0.549800796812749, 'number': 244}, 'P': {'precision': 0.7327823691460055, 'recall': 0.6751269035532995, 'f1': 0.702774108322325, 'number': 394}, 'overall_precision': 0.6505636070853462, 'overall_recall': 0.6332288401253918, 'overall_f1': 0.6417791898332009, 'overall_accuracy': 0.7852555869509376}
			------------EPOCH 22---------------
Loss:  tensor(19.0113, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.8349, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.3351, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.8676, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.3807, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.3589, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.1281, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.6936, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.1540, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.4816, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.8429, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.2645, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.9247, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.3680, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.0290, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.5968, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.1484, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.2551, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.4653, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.5631, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.6153, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.6100, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.0700, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.5940, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.8397, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.5171, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.0445, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.2498, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.2569, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.5337, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.1575, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.1427, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.1136, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.7722, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.8264, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.5707, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.6636, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.6078, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.8475, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.8505, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.3740, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.2122, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.8674, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.0789, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.0299, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.9996, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.5694, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.6323, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.6434, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.2730, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.1145, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.2350, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.0610, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.7047, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.5283, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.9303, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.3913, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.1055, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5328185328185329, 'recall': 0.5655737704918032, 'f1': 0.5487077534791253, 'number': 244}, 'P': {'precision': 0.7327823691460055, 'recall': 0.6751269035532995, 'f1': 0.702774108322325, 'number': 394}, 'overall_precision': 0.6495176848874598, 'overall_recall': 0.6332288401253918, 'overall_f1': 0.6412698412698412, 'overall_accuracy': 0.7859619830464937}
			------------EPOCH 23---------------
Loss:  tensor(17.6454, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.5227, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.1663, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.4215, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.3696, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.6996, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.4166, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.4820, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.0228, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.2519, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.3143, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.8770, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.3663, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.8593, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.7183, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.7006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.5965, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.3876, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.0323, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.6230, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.1597, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.3394, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.2011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.5042, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.2564, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.9522, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.5539, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.7923, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.0457, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.5031, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.3098, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.4409, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.1728, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.4845, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.0447, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.7248, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.2982, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.2882, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.5588, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.7074, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.9742, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.4582, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.1469, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.2955, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.9553, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.5227, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.1438, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.8459, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.1602, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.0071, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.2577, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.4516, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.2897, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.2400, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.1473, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.8794, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.6662, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.0574, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5328185328185329, 'recall': 0.5655737704918032, 'f1': 0.5487077534791253, 'number': 244}, 'P': {'precision': 0.739010989010989, 'recall': 0.682741116751269, 'f1': 0.7097625329815302, 'number': 394}, 'overall_precision': 0.6532905296950241, 'overall_recall': 0.6379310344827587, 'overall_f1': 0.6455194290245837, 'overall_accuracy': 0.785897765219625}
			------------EPOCH 24---------------
Loss:  tensor(16.3417, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.5683, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.1470, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.0313, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.5174, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.9067, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.2580, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.8214, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.7603, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.3513, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.1192, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.5269, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.0604, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.3772, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.5917, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.8277, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.7066, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.5864, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.8181, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.9465, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.0435, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.8762, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.5536, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.7457, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.6183, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.7814, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.4154, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.9578, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.8750, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.6216, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.3136, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.9150, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.3591, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.6138, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.6022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.9969, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.5366, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.2687, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.7949, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.3855, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.8990, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.7966, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.0485, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.7256, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.2766, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.1588, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.8568, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.3729, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.7713, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.4494, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.5777, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.7539, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.9821, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.4510, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.8468, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.0320, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.0576, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.2702, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5287356321839081, 'recall': 0.5655737704918032, 'f1': 0.5465346534653465, 'number': 244}, 'P': {'precision': 0.7417582417582418, 'recall': 0.6852791878172588, 'f1': 0.7124010554089709, 'number': 394}, 'overall_precision': 0.6528, 'overall_recall': 0.6394984326018809, 'overall_f1': 0.6460807600950119, 'overall_accuracy': 0.7870536861032623}
			------------EPOCH 25---------------
Loss:  tensor(15.2939, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.8059, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.2335, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.6553, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.9844, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.2277, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.5726, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.4134, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.6176, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.7949, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.3507, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.2156, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.9104, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.9193, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.7360, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.1054, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.3279, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.9354, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.9297, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.5642, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.1658, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.4783, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.9807, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.3260, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.0485, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.0698, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.4760, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.1690, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.0655, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.2667, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.6270, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.4860, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.7684, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.8815, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.2037, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.4490, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.7357, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.2147, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.4751, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.3101, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.9880, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.2646, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.0981, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.2573, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.9883, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.2463, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.7237, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.8934, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.5496, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.3638, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.4312, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.2162, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.2371, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.3738, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.7303, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.5297, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.8949, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.4079, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5384615384615384, 'recall': 0.6024590163934426, 'f1': 0.5686653771760154, 'number': 244}, 'P': {'precision': 0.717391304347826, 'recall': 0.6700507614213198, 'f1': 0.6929133858267716, 'number': 394}, 'overall_precision': 0.641185647425897, 'overall_recall': 0.6442006269592476, 'overall_f1': 0.6426896012509772, 'overall_accuracy': 0.7909067557153866}
			------------EPOCH 26---------------
Loss:  tensor(16.7879, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.6388, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.0936, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.7874, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.7277, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.1286, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.6971, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.7160, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.8428, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.0877, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.9108, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.6092, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.1161, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.6764, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.6729, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.0687, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.6538, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.3998, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.7196, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.1775, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.7792, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.4841, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.7531, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.7186, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.7769, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.6205, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(78.1268, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.0679, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.4005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.5336, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.8016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.8442, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.5517, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.4991, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.2219, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.1139, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.1399, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.4115, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.9474, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.9425, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.3286, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.9783, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.2787, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.0285, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.6355, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.2250, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.8558, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.8648, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.8023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.9020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.5058, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.2688, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.5115, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.1238, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.2576, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.5292, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.8167, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.8806, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5427509293680297, 'recall': 0.5983606557377049, 'f1': 0.5692007797270955, 'number': 244}, 'P': {'precision': 0.7174515235457064, 'recall': 0.6573604060913706, 'f1': 0.6860927152317882, 'number': 394}, 'overall_precision': 0.6428571428571429, 'overall_recall': 0.6347962382445141, 'overall_f1': 0.6388012618296531, 'overall_accuracy': 0.786861032622656}
			------------EPOCH 27---------------
Loss:  tensor(17.5232, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.5416, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.9310, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.5459, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.2177, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.7804, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.1268, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.4845, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.4929, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.0292, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.7446, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.1511, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.8397, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.5925, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.1339, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.9282, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.1387, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.3525, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.7391, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.3623, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.4072, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.5049, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.1680, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.9044, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.7235, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.3735, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.9413, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.3916, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.1099, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.5652, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.4806, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.0716, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.2676, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.9440, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.7760, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.0764, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.1063, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.8641, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.9524, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.4787, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.0699, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.1856, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.3241, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.2114, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.4252, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.1170, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.9403, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.4675, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.1486, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.4060, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.9711, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.1381, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.3953, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.0899, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.9849, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.8160, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.7507, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.6280, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5756302521008403, 'recall': 0.5614754098360656, 'f1': 0.5684647302904565, 'number': 244}, 'P': {'precision': 0.701530612244898, 'recall': 0.6979695431472082, 'f1': 0.6997455470737914, 'number': 394}, 'overall_precision': 0.653968253968254, 'overall_recall': 0.64576802507837, 'overall_f1': 0.6498422712933755, 'overall_accuracy': 0.7928975083483175}
			------------EPOCH 28---------------
Loss:  tensor(17.5222, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.6574, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.2581, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.6803, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.9841, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.1977, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.3538, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.1861, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.9467, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.9628, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.7705, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.3226, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.8578, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.9262, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.9419, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.2429, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.7901, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.5776, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.5551, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.7786, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.3934, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.8753, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.9363, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.8546, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.4836, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.8949, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.7679, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.8595, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.0877, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.1380, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.0423, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.6983, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.0618, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.2914, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.0500, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.1454, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.8516, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.4848, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.5822, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.4672, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.5648, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.1190, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.7141, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.2858, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.1267, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.3767, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.3104, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.7888, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.7600, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.9190, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.9047, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.4533, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.3835, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.5977, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.3158, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.6689, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.1448, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.8884, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5932203389830508, 'recall': 0.5737704918032787, 'f1': 0.5833333333333334, 'number': 244}, 'P': {'precision': 0.7142857142857143, 'recall': 0.6979695431472082, 'f1': 0.7060333761232349, 'number': 394}, 'overall_precision': 0.6682769726247987, 'overall_recall': 0.6504702194357367, 'overall_f1': 0.6592533756949961, 'overall_accuracy': 0.7965579244798356}
			------------EPOCH 29---------------
Loss:  tensor(15.4133, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.4948, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.7272, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.6503, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.0308, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.3721, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.1681, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.0733, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.2047, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.4224, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.2009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.2607, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.1707, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.2584, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.7444, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.7001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.2981, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.7778, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.8819, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.4370, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.9219, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.3759, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.4396, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.6278, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.0001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.5600, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.6997, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.0497, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.4977, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.9834, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.3754, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.9370, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.5129, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.9273, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.8377, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.5291, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.2270, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.3897, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.0715, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.8097, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.5354, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.0197, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.0016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.7050, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.8372, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.9847, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.6041, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.6272, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.2795, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.4761, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.2932, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.7752, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.9963, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.5736, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.7788, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.6459, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.1510, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.7674, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5756302521008403, 'recall': 0.5614754098360656, 'f1': 0.5684647302904565, 'number': 244}, 'P': {'precision': 0.71875, 'recall': 0.700507614213198, 'f1': 0.7095115681233932, 'number': 394}, 'overall_precision': 0.6639871382636656, 'overall_recall': 0.6473354231974922, 'overall_f1': 0.6555555555555557, 'overall_accuracy': 0.7979064988440792}
			------------EPOCH 30---------------
Loss:  tensor(14.2006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.3889, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.3231, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.4053, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.5604, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.5819, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.5298, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.4055, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.2577, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.6160, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.0160, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.3266, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.1097, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.2846, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.2353, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.0485, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.0537, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.2174, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.0755, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.9839, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.0983, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.0046, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.8937, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.5317, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.6782, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.8579, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.3328, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.4351, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.3892, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.7899, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.7724, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.3885, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.0634, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.4676, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.8701, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.1935, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.2720, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.6232, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.0502, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.4312, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.0663, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.6461, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.9744, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.3080, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.2240, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.6689, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.6095, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.1676, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.1142, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.0811, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.6813, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.5693, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.1148, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.8241, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.1992, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.8769, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.4211, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.1294, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5447154471544715, 'recall': 0.5491803278688525, 'f1': 0.5469387755102041, 'number': 244}, 'P': {'precision': 0.710594315245478, 'recall': 0.6979695431472082, 'f1': 0.704225352112676, 'number': 394}, 'overall_precision': 0.6461295418641391, 'overall_recall': 0.6410658307210031, 'overall_f1': 0.6435877261998426, 'overall_accuracy': 0.7922553300796301}


		-------------RUN 3-----------
			------------EPOCH 1---------------
Loss:  tensor(1931.4139, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2774.9736, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2229.4277, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2004.2278, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1635.0752, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2710.9941, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(645.2933, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1131.3936, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1405.1421, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(826.2828, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2418.3923, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2497.8906, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1526.8169, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2081.1196, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2936.9172, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2369.6577, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3575.0278, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3071.6885, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1568.8623, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2472.8467, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2950.8047, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2794.3979, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2849.7661, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1527.5571, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1730.3145, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1714.6179, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1570.0991, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(832.7328, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1600.8878, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2091.2656, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2043.0132, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1497.3843, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2691.7671, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3867.7004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2355.1958, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2152.1152, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1960.5151, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1884.3644, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1778.0659, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2549.2319, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1546.8650, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1948.9718, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1389.9448, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1305.5490, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(917.6121, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2031.8684, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1381.2347, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1190.6095, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1016.9621, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(659.6531, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1708.8212, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2097.7349, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1678.9392, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1793.7102, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1581.7760, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1337.8618, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2186.6624, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1177.6577, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2098.1348, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.06363636363636363, 'recall': 0.05668016194331984, 'f1': 0.05995717344753747, 'number': 247}, 'P': {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'number': 275}, 'overall_precision': 0.04827586206896552, 'overall_recall': 0.02681992337164751, 'overall_f1': 0.034482758620689655, 'overall_accuracy': 0.3551483420593368}
			------------EPOCH 2---------------
Loss:  tensor(1490.0090, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1868.7698, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1487.8271, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1354.3859, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1178.8409, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2103.9287, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(458.0067, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(818.3062, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1073.3860, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(632.0048, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1779.6040, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1852.0676, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1248.1274, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1477.3245, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2117.8616, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1582.3926, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2809.8467, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2272.5044, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1174.4199, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2282.1169, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2485.5347, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2400.6421, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2412.1133, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1220.7996, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1530.0698, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1436.7737, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1310.6887, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(726.4827, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1216.8336, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1717.2367, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1466.7739, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1159.5391, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2258.4536, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3352.4421, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1932.9352, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1747.3904, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1524.2614, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1759.7701, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1591.0444, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2185.1978, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1325.9196, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1716.2386, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1248.3062, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1118.6831, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(813.2971, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1727.9611, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1063.2019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(948.2019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(826.1743, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(541.8077, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1315.1262, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1559.6628, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1109.8528, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1323.3101, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1168.8672, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1045.2649, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1932.2615, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(930.6103, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1670.2620, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.14057507987220447, 'recall': 0.17813765182186234, 'f1': 0.15714285714285717, 'number': 247}, 'P': {'precision': 0.2168141592920354, 'recall': 0.1781818181818182, 'f1': 0.19560878243512972, 'number': 275}, 'overall_precision': 0.1725417439703154, 'overall_recall': 0.1781609195402299, 'overall_f1': 0.17530631479736097, 'overall_accuracy': 0.5420302501454334}
			------------EPOCH 3---------------
Loss:  tensor(1379.1143, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1585.9725, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1242.5333, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1106.0496, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(976.3512, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1952.3440, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(338.2494, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(706.5896, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(949.4421, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(469.7141, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1485.1389, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1555.2361, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1080.4547, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1313.8938, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1635.8760, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1274.7211, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2277.2603, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1803.3566, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(931.0584, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1697.5582, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1955.5934, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1930.2421, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1677.3665, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(948.3312, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1226.1307, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1189.7207, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1210.6161, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(627.9336, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1024.0305, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1474.5094, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1291.4873, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(962.4585, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1888.9749, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2969.7339, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1609.6714, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1479.9033, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1080.6545, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1399.7563, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1321.5530, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1807.2206, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(999.7451, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1410.7893, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(959.3303, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(903.1870, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(610.8530, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1344.4994, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(977.4788, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(864.6274, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(750.3136, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(497.8659, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1195.5808, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1390.9337, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1053.5767, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1141.7781, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1013.9583, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(903.2913, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1565.5356, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(736.7688, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1346.2522, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.30357142857142855, 'recall': 0.3441295546558704, 'f1': 0.32258064516129026, 'number': 247}, 'P': {'precision': 0.4072948328267477, 'recall': 0.48727272727272725, 'f1': 0.4437086092715232, 'number': 275}, 'overall_precision': 0.35960591133004927, 'overall_recall': 0.41954022988505746, 'overall_f1': 0.38726790450928383, 'overall_accuracy': 0.622454915648633}
			------------EPOCH 4---------------
Loss:  tensor(1051.0388, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1181.5801, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(886.8699, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(860.2391, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(819.2316, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1401.8748, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(279.7910, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(537.6765, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(842.3960, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(384.0175, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1200.4976, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1299.7883, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(927.2797, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1031.4714, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1368.1663, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(866.2319, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1746.0332, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1399.8423, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(609.0387, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1181.6479, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1572.5448, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1625.2148, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1501.8083, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(821.8408, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1019.6671, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1062.0884, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(985.3998, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(480.5608, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(799.9087, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1102.9258, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(933.1533, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(717.9559, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1589.7195, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2296.7961, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1175.7051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1029.0413, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(711.6920, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1096.7069, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1036.5869, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1519.9418, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(746.5833, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1211.2885, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(717.6722, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(736.9326, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(535.2704, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1059.5676, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(708.2060, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(650.2002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(573.8192, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(386.0344, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(885.0092, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1029.7936, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(655.7314, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(945.2614, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(842.2094, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(726.0634, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1162.5790, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(619.9960, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1081.3699, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3180428134556575, 'recall': 0.42105263157894735, 'f1': 0.36236933797909415, 'number': 247}, 'P': {'precision': 0.436426116838488, 'recall': 0.4618181818181818, 'f1': 0.4487632508833922, 'number': 275}, 'overall_precision': 0.3737864077669903, 'overall_recall': 0.4425287356321839, 'overall_f1': 0.4052631578947368, 'overall_accuracy': 0.6751018033740547}
			------------EPOCH 5---------------
Loss:  tensor(816.8496, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(866.5251, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(668.5299, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(683.1986, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(631.5137, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(925.2816, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(220.8711, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(473.8617, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(747.5725, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(275.5980, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1014.7964, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1108.7937, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(801.1656, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(893.7930, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1128.9203, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(723.9424, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1439.8967, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1192.9727, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(532.3964, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1253.2898, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1317.4250, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1266.6465, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1543.4390, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(619.5359, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(709.0809, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(863.1519, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(759.1125, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(329.8271, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(801.6056, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1093.4978, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1004.0146, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(741.0547, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1305.2935, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3176.3704, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1488.4502, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1132.4026, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(482.7665, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(670.0126, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(716.7137, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(931.4575, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(558.8957, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1058.3510, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(640.5485, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(660.3887, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(594.1027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1371.4028, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(726.9252, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(627.8461, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(598.5194, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(469.3228, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1023.2126, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1251.1550, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(622.5283, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(870.6630, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(758.6801, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(992.6636, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1547.4348, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(706.9413, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1271.2297, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4187725631768953, 'recall': 0.46963562753036436, 'f1': 0.44274809160305345, 'number': 247}, 'P': {'precision': 0.35892116182572614, 'recall': 0.6290909090909091, 'f1': 0.4570673712021136, 'number': 275}, 'overall_precision': 0.380764163372859, 'overall_recall': 0.553639846743295, 'overall_f1': 0.45120999219359875, 'overall_accuracy': 0.6235456660849331}
			------------EPOCH 6---------------
Loss:  tensor(1354.7683, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1309.2245, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(690.1744, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(879.9592, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(824.9031, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1395.8478, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(238.6228, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(416.9220, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(678.6458, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(280.6104, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1008.0278, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1038.7577, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(710.0150, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(881.7607, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1005.9909, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(803.2278, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1359.4585, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1253.7678, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(579.1594, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1107.6796, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1389.1384, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1284.4221, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1400.6277, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(695.4103, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(731.7440, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(883.8357, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(923.2993, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(393.4794, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(692.5585, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1109.1957, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(815.5078, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(603.4051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1194.1516, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2122.3496, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1001.5743, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(757.9335, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(458.1375, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(725.9054, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(755.5592, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(896.6760, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(498.1557, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(933.0895, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(530.9283, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(591.2349, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(336.9951, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(859.6247, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(701.8575, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(369.5768, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(251.6727, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(332.5943, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(532.1928, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(682.2135, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(332.5626, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(496.0346, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(433.2744, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(454.9035, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(691.7485, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(537.4241, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(930.9077, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4690265486725664, 'recall': 0.2145748987854251, 'f1': 0.29444444444444445, 'number': 247}, 'P': {'precision': 0.3824130879345603, 'recall': 0.68, 'f1': 0.48952879581151837, 'number': 275}, 'overall_precision': 0.39867109634551495, 'overall_recall': 0.45977011494252873, 'overall_f1': 0.4270462633451957, 'overall_accuracy': 0.6196189645142525}
			------------EPOCH 7---------------
Loss:  tensor(1131.6257, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1243.9758, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1207.0751, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(978.4890, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(407.8370, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(818.0431, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(166.3658, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(426.6936, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(526.1389, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(153.0928, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(648.2286, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(681.7482, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(628.0530, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(900.0162, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(914.3024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(715.5151, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1379.9825, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1210.4751, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(576.7604, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(760.0700, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1006.4390, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(911.5117, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(986.7062, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(708.2936, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(451.4142, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(607.3857, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(585.2162, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(262.1809, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(518.0898, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(791.3732, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(545.1941, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(411.0685, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1084.7098, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2058.9121, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(795.9440, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(681.1946, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(472.3134, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(768.8381, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(801.1348, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1077.1062, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(689.4496, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1243.1465, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(693.5729, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(734.1391, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(452.4418, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(792.9148, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(499.9586, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(412.4722, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(274.4216, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(253.3208, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(501.5712, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(639.4269, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(302.1591, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(381.2891, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(390.5426, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(371.6955, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(768.2532, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(413.7811, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(657.1750, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3619631901840491, 'recall': 0.7165991902834008, 'f1': 0.48097826086956524, 'number': 247}, 'P': {'precision': 0.5833333333333334, 'recall': 0.3563636363636364, 'f1': 0.4424379232505644, 'number': 275}, 'overall_precision': 0.4185692541856925, 'overall_recall': 0.5268199233716475, 'overall_f1': 0.46649703138252757, 'overall_accuracy': 0.6307446189645143}
			------------EPOCH 8---------------
Loss:  tensor(527.8287, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(652.8771, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(581.4886, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(521.5793, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(563.4129, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(689.0914, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(160.4955, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(341.2987, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(490.6165, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(126.9963, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(587.0594, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(529.4514, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(351.2153, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(538.3831, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(619.2371, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(422.0669, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(653.1639, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(731.3674, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(240.9612, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(434.7152, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(689.7316, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(602.6514, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(768.9742, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(323.3118, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(348.9171, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(383.4984, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(405.9373, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(198.4920, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(347.0988, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(535.1902, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(308.9953, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(261.1078, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(582.7501, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1159.1514, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(514.5851, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(296.1471, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(183.9425, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(303.3311, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(360.5296, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(395.4089, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(247.5252, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(697.6109, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(362.4532, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(403.4214, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(268.8219, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(540.1479, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(252.4505, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(178.1967, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(148.1059, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(171.1897, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(209.4552, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(225.9391, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(163.9919, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(195.0206, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(209.6147, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(246.7022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(366.1718, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(254.7325, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(387.9276, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.44281524926686217, 'recall': 0.611336032388664, 'f1': 0.5136054421768708, 'number': 247}, 'P': {'precision': 0.5269709543568465, 'recall': 0.4618181818181818, 'f1': 0.49224806201550386, 'number': 275}, 'overall_precision': 0.47766323024054985, 'overall_recall': 0.5325670498084292, 'overall_f1': 0.5036231884057971, 'overall_accuracy': 0.68717277486911}
			------------EPOCH 9---------------
Loss:  tensor(292.8297, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(400.2576, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(326.0067, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(326.7749, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(231.7925, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(286.4240, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(74.9050, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(179.9820, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(283.0832, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(78.8400, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(295.3356, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(267.2423, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(217.6235, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(386.7526, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(428.5270, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(286.2843, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(408.4772, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(471.0403, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(145.1934, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(277.0977, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(444.9899, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(372.7215, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(465.2642, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(235.5031, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(213.4725, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(244.9457, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(255.3510, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(170.6462, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(221.3609, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(377.9128, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(181.1261, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(164.1524, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(403.4201, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(676.8600, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(211.3815, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(151.3631, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(140.7571, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(203.4099, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(210.5427, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(246.9963, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(136.6409, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(403.1139, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(229.1443, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(257.9636, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(134.3943, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(349.0461, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(131.9043, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(93.1045, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(100.5785, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(106.3883, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(134.6497, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(129.6747, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(86.7163, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(118.0274, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(130.3539, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(159.0399, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(162.8864, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(163.2559, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(224.1488, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.40808823529411764, 'recall': 0.4493927125506073, 'f1': 0.4277456647398844, 'number': 247}, 'P': {'precision': 0.47333333333333333, 'recall': 0.5163636363636364, 'f1': 0.49391304347826087, 'number': 275}, 'overall_precision': 0.4423076923076923, 'overall_recall': 0.4846743295019157, 'overall_f1': 0.4625228519195612, 'overall_accuracy': 0.6900087260034904}
			------------EPOCH 10---------------
Loss:  tensor(204.1138, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(231.4760, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(166.7157, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(176.9361, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(117.0985, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(219.4664, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.0241, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(131.9539, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(208.7166, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.4977, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(241.9962, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(215.3903, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(174.0370, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(297.5152, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(309.6385, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(277.6040, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(297.7331, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(382.8385, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(116.6008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(202.4836, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(162.9796, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(164.3403, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(266.0515, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(111.6702, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(97.9522, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(109.8143, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(151.2725, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(84.2241, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(135.7898, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(217.4945, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(99.5467, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(93.1620, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(188.0727, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(419.1304, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(83.0462, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(101.3009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(146.6215, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(199.3974, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(136.3148, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(179.7535, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(126.5213, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(269.8139, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(151.1685, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(156.5214, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(100.3270, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(265.9281, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(102.1160, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(69.8155, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(100.4092, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(77.1160, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(144.5393, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(155.6280, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(74.0351, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(76.5785, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(98.9189, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(122.9151, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(85.4240, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(135.9367, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(115.0419, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5178571428571429, 'recall': 0.46963562753036436, 'f1': 0.4925690021231423, 'number': 247}, 'P': {'precision': 0.5228426395939086, 'recall': 0.7490909090909091, 'f1': 0.6158445440956652, 'number': 275}, 'overall_precision': 0.5210355987055016, 'overall_recall': 0.6168582375478927, 'overall_f1': 0.5649122807017544, 'overall_accuracy': 0.7359656777196044}
			------------EPOCH 11---------------
Loss:  tensor(130.4229, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(114.5771, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(61.6221, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(83.7525, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(75.9412, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(101.8529, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.1198, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(81.2017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(115.1612, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.4530, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(117.2758, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(126.7159, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(97.2765, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(179.2534, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(181.8472, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(182.5275, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(215.0975, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(315.1013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(78.0170, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(147.3654, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(197.5229, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(117.4211, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(161.7878, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(144.4051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(154.7223, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(145.0677, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(140.7641, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(92.6389, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(106.6278, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(199.1117, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(88.6618, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(99.3047, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(136.5488, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(249.7100, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.9794, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.1128, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(74.6627, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(88.9064, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(89.5007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(86.7029, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.8520, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(155.3095, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(115.1716, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(74.8015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(66.0809, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(231.9221, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(78.4436, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(59.8821, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.5469, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.6275, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(109.0912, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(85.3370, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.7258, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(104.4544, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(140.8923, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(137.5409, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(74.6294, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(93.3509, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(100.8841, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.569060773480663, 'recall': 0.41700404858299595, 'f1': 0.4813084112149532, 'number': 247}, 'P': {'precision': 0.5047619047619047, 'recall': 0.7709090909090909, 'f1': 0.6100719424460431, 'number': 275}, 'overall_precision': 0.5241264559068219, 'overall_recall': 0.603448275862069, 'overall_f1': 0.5609973285841496, 'overall_accuracy': 0.6994618964514252}
			------------EPOCH 12---------------
Loss:  tensor(135.5037, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(97.3201, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.6092, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(98.9401, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(66.2422, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(119.6301, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.6588, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(78.2264, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(62.0961, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.4692, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(112.3906, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(105.8437, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(84.4663, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(153.7121, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(186.7857, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(97.5369, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(132.2960, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(214.6416, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.7821, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(146.7228, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(97.3553, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(86.7537, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(115.9800, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.0871, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(69.0288, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(73.8796, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(106.5014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(71.3157, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(106.0254, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(137.0644, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(82.4426, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(65.1327, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(206.7953, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(239.6606, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.8915, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(86.0364, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(76.0076, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(66.1334, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(91.0571, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(85.9374, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.4228, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(108.6206, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(94.6206, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.3748, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.6644, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(143.7461, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.9015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.6489, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.6228, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.8411, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(99.1290, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(88.6182, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(84.4224, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(112.8749, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(181.0963, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(161.5906, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(535.0218, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(201.0358, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(291.6625, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5698924731182796, 'recall': 0.2145748987854251, 'f1': 0.31176470588235294, 'number': 247}, 'P': {'precision': 0.45714285714285713, 'recall': 0.8145454545454546, 'f1': 0.5856209150326797, 'number': 275}, 'overall_precision': 0.4751286449399657, 'overall_recall': 0.5306513409961686, 'overall_f1': 0.5013574660633485, 'overall_accuracy': 0.6625945317044794}
			------------EPOCH 13---------------
Loss:  tensor(256.2026, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(178.5797, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(118.8122, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(183.9714, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.3970, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.8198, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.8604, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(82.1410, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.8882, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.7304, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(81.8483, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(131.3797, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(80.1503, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(176.8420, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(129.1848, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(113.5583, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(348.1151, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(399.1828, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(211.5370, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(176.9853, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(259.8223, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(310.0859, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(388.6426, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(193.9315, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(73.0238, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(69.7764, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(81.6308, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(59.0203, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(66.6837, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(91.4639, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.4073, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.4851, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(108.5896, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(269.1397, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.6666, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.5830, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(67.9598, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(75.9514, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.7506, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(74.3026, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.5983, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(83.6875, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(78.8239, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.6581, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(79.0755, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(127.0581, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.0975, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.3428, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.9874, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.9250, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(79.7027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.4962, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.1036, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.6539, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.1243, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(61.6102, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(61.6666, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.3150, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.5406, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4845360824742268, 'recall': 0.1902834008097166, 'f1': 0.27325581395348836, 'number': 247}, 'P': {'precision': 0.4560327198364008, 'recall': 0.8109090909090909, 'f1': 0.5837696335078534, 'number': 275}, 'overall_precision': 0.46075085324232085, 'overall_recall': 0.5172413793103449, 'overall_f1': 0.48736462093862815, 'overall_accuracy': 0.6509598603839442}
			------------EPOCH 14---------------
Loss:  tensor(63.3202, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(66.4576, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.9997, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(67.7578, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(223.7606, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(529.1545, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(114.5656, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(85.5034, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(685.8552, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(716.1623, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1109.0227, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1638.5546, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(79.4047, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(111.4809, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(162.2843, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(138.6010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(133.4847, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(261.1403, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.7585, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(98.8631, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(840.2195, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1138.4913, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(194.3057, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(919.8773, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1055.7156, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1036.4014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(462.8202, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(757.3779, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(181.5226, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(472.0338, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(216.1245, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(246.9793, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(214.7059, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(310.1102, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(120.1937, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(85.8589, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(80.4222, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(131.2652, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(160.7301, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(147.0797, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(126.6906, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(386.9785, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(168.1314, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(174.9286, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(177.3311, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(397.2534, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(197.3604, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(100.5930, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(114.4565, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(249.0459, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(401.5093, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(212.0794, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(84.6758, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(190.8585, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(191.6216, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(185.2127, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(239.5412, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(161.9426, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(262.2530, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.6041666666666666, 'recall': 0.23481781376518218, 'f1': 0.33819241982507287, 'number': 247}, 'P': {'precision': 0.45901639344262296, 'recall': 0.8145454545454546, 'f1': 0.5871559633027523, 'number': 275}, 'overall_precision': 0.4828767123287671, 'overall_recall': 0.5402298850574713, 'overall_f1': 0.5099457504520796, 'overall_accuracy': 0.6721204188481675}
			------------EPOCH 15---------------
Loss:  tensor(343.0414, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(334.2269, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(386.7150, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(286.7268, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(163.0613, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(232.7101, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.4876, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(192.3335, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(165.6806, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.9252, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(163.8204, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(136.3411, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(243.0558, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(394.9311, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(728.9002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(149.1209, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(350.4508, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(356.1566, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(79.4009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(279.7235, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(251.4062, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(322.2192, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(210.6206, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(169.3654, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(136.7436, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(128.7686, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(139.0768, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(76.8413, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(230.3995, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(451.5840, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(207.3006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(149.3050, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(730.5267, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1005.7326, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(353.0825, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(342.6791, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(345.2781, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(371.4116, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1056.7336, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1072.3633, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(248.8546, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(453.4376, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(289.7926, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(244.7973, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(72.6273, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(224.1456, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(173.5984, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.3515, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.3824, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.7862, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.0601, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.7456, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.6146, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.8896, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.0217, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(68.8214, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(59.6205, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(61.5524, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(80.7056, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.573170731707317, 'recall': 0.1902834008097166, 'f1': 0.2857142857142857, 'number': 247}, 'P': {'precision': 0.4444444444444444, 'recall': 0.8872727272727273, 'f1': 0.5922330097087378, 'number': 275}, 'overall_precision': 0.4611727416798732, 'overall_recall': 0.5574712643678161, 'overall_f1': 0.5047701647875109, 'overall_accuracy': 0.6431791739383362}
			------------EPOCH 16---------------
Loss:  tensor(362.3325, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(127.1805, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(244.3341, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(168.0461, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(217.6977, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(298.4410, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.6746, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(211.5999, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(517.0009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(238.0923, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(336.9637, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(214.8356, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(423.7422, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(654.4924, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1236.5275, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(406.7368, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(714.6325, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(518.2261, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(233.7854, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(687.3785, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(445.6166, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(707.0872, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(309.0724, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(389.7976, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(110.6414, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(150.0216, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(136.4021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(105.0751, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(78.5338, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(133.7150, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(77.1464, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.5330, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(172.4929, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(288.7640, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(101.7480, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(87.6328, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(234.0633, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(193.5847, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(622.3177, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(484.6171, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(356.6400, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(643.8911, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(468.9270, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(396.3609, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(187.1600, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(704.1327, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(857.8748, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(664.1306, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(252.8316, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(437.8945, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(985.7899, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1350.5603, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(693.5144, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(451.4769, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(504.6851, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(238.1306, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(361.2091, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(278.7979, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(285.5311, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4315068493150685, 'recall': 0.5101214574898786, 'f1': 0.4675324675324675, 'number': 247}, 'P': {'precision': 0.5391849529780565, 'recall': 0.6254545454545455, 'f1': 0.5791245791245792, 'number': 275}, 'overall_precision': 0.4877250409165303, 'overall_recall': 0.5708812260536399, 'overall_f1': 0.5260370697263902, 'overall_accuracy': 0.720913321698662}
			------------EPOCH 17---------------
Loss:  tensor(118.0840, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(170.0243, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(93.7774, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(87.4913, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(133.3167, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(88.4296, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(72.2200, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(185.0393, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(98.0890, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.5934, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(109.3289, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(112.4696, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(125.5499, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(206.7466, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(375.8713, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(113.6926, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(434.0757, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(361.4041, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(109.4727, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(359.7303, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(639.9205, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(604.4060, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(346.9922, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(364.5583, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(301.7237, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(613.7930, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(727.3305, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(203.7663, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(422.3076, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(842.2811, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(772.9310, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(583.8910, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(670.3949, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1437.6052, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(454.9495, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(368.6986, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(94.5840, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(218.5519, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(152.6099, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(115.9240, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(109.4269, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(155.9888, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(126.9300, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(104.9807, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(75.4209, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(128.6169, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(101.5646, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.9348, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.0125, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(157.3212, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(115.0965, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(99.3171, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(169.6537, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(102.6459, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(139.9942, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(138.9100, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(344.1331, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(214.4728, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(271.7606, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3758241758241758, 'recall': 0.6923076923076923, 'f1': 0.4871794871794872, 'number': 247}, 'P': {'precision': 0.6047904191616766, 'recall': 0.36727272727272725, 'f1': 0.45701357466063347, 'number': 275}, 'overall_precision': 0.43729903536977494, 'overall_recall': 0.5210727969348659, 'overall_f1': 0.4755244755244755, 'overall_accuracy': 0.6607766143106457}
			------------EPOCH 18---------------
Loss:  tensor(155.8099, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(276.8622, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(181.8251, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(217.7111, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(248.4851, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(239.5770, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(99.7615, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(319.2417, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(357.7343, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(68.1730, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(584.3152, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(600.9374, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(215.1605, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(377.2809, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(249.3391, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(355.6113, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(224.1360, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(287.9714, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(98.4710, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(125.5919, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(133.6379, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(94.2363, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(109.0100, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.8153, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.0849, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.7057, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.7708, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(61.9945, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(68.1574, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(88.0262, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.8752, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.3792, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(73.2229, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(159.9942, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.3334, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.3566, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.5876, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(126.1668, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.4684, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.4597, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(73.5982, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(108.6646, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(71.9923, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.2282, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.6015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(70.7437, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.3563, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.8953, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.5674, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.3815, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(72.4001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.4048, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.9108, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.9455, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.9437, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.0126, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.8383, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.1106, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.8736, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4720812182741117, 'recall': 0.3765182186234818, 'f1': 0.4189189189189189, 'number': 247}, 'P': {'precision': 0.5134474327628362, 'recall': 0.7636363636363637, 'f1': 0.6140350877192983, 'number': 275}, 'overall_precision': 0.5, 'overall_recall': 0.5804597701149425, 'overall_f1': 0.5372340425531915, 'overall_accuracy': 0.705933682373473}
			------------EPOCH 19---------------
Loss:  tensor(42.4218, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.1713, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.1428, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.1154, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.0315, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.3750, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.0576, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.6739, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(66.5276, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.2552, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.9773, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.0889, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.7682, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(87.0038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(118.8940, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.2735, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(69.9574, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(129.6096, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.0891, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(70.5492, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.7335, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(73.6367, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(59.0949, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.4340, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.0397, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(65.7943, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.7223, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.8619, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(78.1330, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(88.8087, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(59.3416, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.5253, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(65.6639, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(202.7052, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.1675, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.7462, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.1291, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.2181, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.3856, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.6691, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.5728, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(66.7148, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.6685, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.7902, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.9859, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.0226, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.2371, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.6035, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.3144, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.9687, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.5018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.3990, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.5034, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.2085, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.6118, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.9409, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.1215, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.3267, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.8983, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.49193548387096775, 'recall': 0.4939271255060729, 'f1': 0.4929292929292929, 'number': 247}, 'P': {'precision': 0.577259475218659, 'recall': 0.72, 'f1': 0.6407766990291262, 'number': 275}, 'overall_precision': 0.5414551607445008, 'overall_recall': 0.6130268199233716, 'overall_f1': 0.5750224618149147, 'overall_accuracy': 0.7308755090168703}
			------------EPOCH 20---------------
Loss:  tensor(26.7627, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.1456, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.5627, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.2525, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.9725, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.0541, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.5008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.1328, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.0354, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.6784, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.4486, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.7756, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.8131, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.1827, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.9235, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.9232, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.5087, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(110.5817, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.6634, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.3054, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.6436, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.8623, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.4642, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.8005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.5773, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.5533, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.7000, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.8646, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.0816, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.9310, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.0040, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.7166, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.0665, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(71.4071, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.5679, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.3252, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.9602, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.6158, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.2418, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.7343, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.2359, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.4616, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.0022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.3734, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.8173, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.4327, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.0362, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.6370, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.7358, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.8605, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.9875, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.4093, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.8114, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.5498, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.5543, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.3076, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.0070, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.2810, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.3863, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5036496350364964, 'recall': 0.5587044534412956, 'f1': 0.5297504798464492, 'number': 247}, 'P': {'precision': 0.6114649681528662, 'recall': 0.6981818181818182, 'f1': 0.6519524617996605, 'number': 275}, 'overall_precision': 0.5612244897959183, 'overall_recall': 0.632183908045977, 'overall_f1': 0.5945945945945946, 'overall_accuracy': 0.7425828970331588}
			------------EPOCH 21---------------
Loss:  tensor(21.3880, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.4175, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.3335, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.2386, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.6636, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.9787, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.5814, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.1295, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.7391, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.9620, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.7079, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.4358, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.1103, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.8700, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.0826, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.2750, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.3412, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(91.4626, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.0529, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.8874, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.1163, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.6137, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.8205, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.2931, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.0106, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.8338, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.4585, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.0446, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.5942, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.6875, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.4441, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.7751, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.3111, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.1184, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.9978, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.4512, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.2063, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.2568, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.7687, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.0449, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.3844, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.0395, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.6883, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.0814, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.7960, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.7045, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.5841, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.7849, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.7814, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.2981, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.1211, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.5759, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.4611, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.1647, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.9407, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.0901, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.7579, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.2042, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.4987, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5055762081784386, 'recall': 0.5506072874493927, 'f1': 0.5271317829457364, 'number': 247}, 'P': {'precision': 0.616822429906542, 'recall': 0.72, 'f1': 0.6644295302013422, 'number': 275}, 'overall_precision': 0.5661016949152542, 'overall_recall': 0.6398467432950191, 'overall_f1': 0.6007194244604317, 'overall_accuracy': 0.7499272833042466}
			------------EPOCH 22---------------
Loss:  tensor(18.4656, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.3597, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.3418, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.4285, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.2661, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.1847, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.9351, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.2218, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.9704, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.5224, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.7690, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.4152, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.2058, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.6596, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.8876, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.5532, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.0239, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(78.9334, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.8963, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.6124, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.2151, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.7267, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.6541, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.9153, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.7035, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.5949, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.3281, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.5437, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.8516, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.1264, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.6316, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.7593, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.9342, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.6808, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.2522, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.2375, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.5138, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.6648, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.1434, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.4193, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.8213, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.0282, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.4881, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.0936, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.6792, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.8992, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.5501, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.2710, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.5324, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.1710, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.5969, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.4120, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.0810, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.1396, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.3311, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.3627, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.4719, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.7063, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.8420, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4756554307116105, 'recall': 0.5141700404858299, 'f1': 0.49416342412451364, 'number': 247}, 'P': {'precision': 0.6006006006006006, 'recall': 0.7272727272727273, 'f1': 0.6578947368421053, 'number': 275}, 'overall_precision': 0.545, 'overall_recall': 0.6264367816091954, 'overall_f1': 0.5828877005347594, 'overall_accuracy': 0.7450552646887726}
			------------EPOCH 23---------------
Loss:  tensor(16.2267, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.3761, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.1654, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.7938, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.9540, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.4931, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.5989, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.9848, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.4447, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.4289, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.5632, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.4119, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.5229, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.5723, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.4003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.3158, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.6400, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(69.0062, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.6757, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.2046, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.7384, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.9212, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.9131, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.7219, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.5539, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.0732, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.1012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.0699, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.1817, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.4022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.3348, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.6835, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.1969, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.9078, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.9437, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.7181, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.9596, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.9999, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.6554, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.4443, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.8248, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.0634, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.0940, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.7124, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.4848, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.2302, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.7351, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.0794, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.3487, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.8493, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.1585, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.0009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.7598, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.9297, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.4689, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.4363, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.3304, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.5618, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.3984, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4981132075471698, 'recall': 0.5344129554655871, 'f1': 0.5156250000000001, 'number': 247}, 'P': {'precision': 0.6134969325153374, 'recall': 0.7272727272727273, 'f1': 0.6655574043261232, 'number': 275}, 'overall_precision': 0.5617597292724196, 'overall_recall': 0.6360153256704981, 'overall_f1': 0.5965858041329739, 'overall_accuracy': 0.7475276323443862}
			------------EPOCH 24---------------
Loss:  tensor(14.3345, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.6857, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.4016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.2768, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.1361, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.5040, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.9604, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.5278, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.5347, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.1511, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.4568, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.9227, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.4874, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.9197, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.1758, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.4270, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.9257, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(61.0485, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.4362, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.5145, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.8953, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.0120, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.8111, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.6069, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.5900, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.2668, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.8886, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.8480, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.0966, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.4200, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.4563, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.4557, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.4998, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.4285, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.7145, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.3955, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.8163, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.9021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.4622, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.5096, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.1670, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.0430, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.4506, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.7498, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.8935, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.0736, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.9898, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.9774, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.4607, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.8389, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.2689, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.9750, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.8407, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.9291, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.9141, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.9727, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.4564, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.3000, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.1711, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.49624060150375937, 'recall': 0.5344129554655871, 'f1': 0.5146198830409356, 'number': 247}, 'P': {'precision': 0.6134969325153374, 'recall': 0.7272727272727273, 'f1': 0.6655574043261232, 'number': 275}, 'overall_precision': 0.5608108108108109, 'overall_recall': 0.6360153256704981, 'overall_f1': 0.5960502692998205, 'overall_accuracy': 0.747673065735893}
			------------EPOCH 25---------------
Loss:  tensor(13.0945, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.1176, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.5624, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.8609, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.0804, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.5235, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5642, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.8526, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.5628, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.2103, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.0660, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.9939, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.5076, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.5223, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.2495, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.1846, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.6248, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.3908, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.3671, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.7163, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.0315, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.7044, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.2907, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.8581, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.7060, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.9931, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.0768, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.4384, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.3832, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.9045, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.4639, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.4328, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.4716, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.1965, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.9791, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.3615, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.9555, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.5272, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.6724, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.8381, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.7080, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.5105, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.1588, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.1252, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.2406, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.1864, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.2486, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.9250, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.6858, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.9334, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.9557, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.8240, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.0935, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.0760, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.2943, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.4409, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.6139, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.8546, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.1656, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.47547169811320755, 'recall': 0.5101214574898786, 'f1': 0.4921875, 'number': 247}, 'P': {'precision': 0.6036585365853658, 'recall': 0.72, 'f1': 0.6567164179104478, 'number': 275}, 'overall_precision': 0.5463743676222597, 'overall_recall': 0.6206896551724138, 'overall_f1': 0.5811659192825112, 'overall_accuracy': 0.7412739965095986}
			------------EPOCH 26---------------
Loss:  tensor(9.9460, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.8013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.9462, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.2006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.3471, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.7330, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2790, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.9519, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.7691, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.1613, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.9079, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.4526, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.0852, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.2120, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.9328, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.6821, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.6548, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.1793, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.2095, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.3773, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.0107, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.4197, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.9921, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.2720, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.9249, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.0479, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.3862, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.2571, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.8296, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.6349, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.9796, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.4956, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.2672, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.5862, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.4050, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.5065, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.9825, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.3039, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.0492, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.7398, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.0769, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.3622, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.1421, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.6151, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.9129, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.8949, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.7554, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.4346, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.1078, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.1314, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.0925, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.4286, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.4860, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.3564, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.3304, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.6508, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.9740, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.8423, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.1434, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5, 'recall': 0.5384615384615384, 'f1': 0.5185185185185186, 'number': 247}, 'P': {'precision': 0.6, 'recall': 0.72, 'f1': 0.6545454545454547, 'number': 275}, 'overall_precision': 0.5553691275167785, 'overall_recall': 0.6340996168582376, 'overall_f1': 0.592128801431127, 'overall_accuracy': 0.7412012798138452}
			------------EPOCH 27---------------
Loss:  tensor(8.3823, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.1637, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.5020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.5009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.0683, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.0081, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.0266, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.1724, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.0386, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.3449, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.9773, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.6619, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.7942, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.0015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.6024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.3194, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.8216, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.9993, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.6970, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.1642, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.8680, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.3338, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.9300, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.7204, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.1054, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.3166, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.6747, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.1979, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.0711, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.3507, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.2713, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.8302, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.5504, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.5615, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.0143, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.8397, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.3766, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.0761, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.5245, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.7987, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.5839, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.4415, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.2590, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.1723, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.7188, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.7451, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.0035, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.4084, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.6095, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.9661, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.9746, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.2200, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.5860, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.7244, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.7305, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.7547, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.2908, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.1913, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.4213, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5130111524163569, 'recall': 0.5587044534412956, 'f1': 0.5348837209302326, 'number': 247}, 'P': {'precision': 0.6042944785276073, 'recall': 0.7163636363636363, 'f1': 0.6555740432612314, 'number': 275}, 'overall_precision': 0.5630252100840336, 'overall_recall': 0.6417624521072797, 'overall_f1': 0.5998209489704566, 'overall_accuracy': 0.7463641652123327}
			------------EPOCH 28---------------
Loss:  tensor(7.7980, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.1196, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.1780, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.9570, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.7409, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.5107, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.9136, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.4486, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.2119, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.4901, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.4130, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.2110, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.9236, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.7281, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.7453, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.7116, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.9701, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.4353, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.0974, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.1850, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.3925, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.2906, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.2814, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.1676, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.5050, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.4045, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.3320, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.7106, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.4275, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.2902, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.2950, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.4017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.2017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.2309, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.7259, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.8411, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.1571, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.8971, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.5617, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.0407, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.1119, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.0857, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.9856, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.4043, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.6818, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.3228, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.9102, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.5651, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.3078, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.1643, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.4173, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.4747, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.2633, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.6827, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.7807, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.8807, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.0520, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.8290, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.3660, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4701492537313433, 'recall': 0.5101214574898786, 'f1': 0.4893203883495146, 'number': 247}, 'P': {'precision': 0.5880597014925373, 'recall': 0.7163636363636363, 'f1': 0.6459016393442623, 'number': 275}, 'overall_precision': 0.5356550580431177, 'overall_recall': 0.6187739463601533, 'overall_f1': 0.5742222222222222, 'overall_accuracy': 0.735892961023851}
			------------EPOCH 29---------------
Loss:  tensor(7.9776, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.8293, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.2403, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.1234, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.2556, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.4841, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.1865, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.6352, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.3470, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.5542, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.4657, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.0805, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.0605, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.6496, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.6986, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.2702, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.6932, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.3277, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.2669, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.1768, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.0486, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.4233, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.0847, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.1271, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.5369, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.9589, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.8744, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.0211, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.9795, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.6558, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.0741, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.0976, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.0508, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.8789, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.3461, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.7718, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.2357, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.6925, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.6522, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.2607, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.7499, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.1266, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.6750, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.2261, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.2936, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.0983, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.8052, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.0135, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.0858, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.0252, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.4217, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.3254, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.2399, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.4910, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.2258, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.8509, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.8361, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.5137, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.4935, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4962686567164179, 'recall': 0.5384615384615384, 'f1': 0.516504854368932, 'number': 247}, 'P': {'precision': 0.5868263473053892, 'recall': 0.7127272727272728, 'f1': 0.6436781609195401, 'number': 275}, 'overall_precision': 0.5465116279069767, 'overall_recall': 0.6302681992337165, 'overall_f1': 0.5854092526690392, 'overall_accuracy': 0.7378563118091914}
			------------EPOCH 30---------------
Loss:  tensor(6.5195, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.5160, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.2307, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.7380, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.5455, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.4911, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.7998, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.3605, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.4065, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.0163, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.8183, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.2947, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.8509, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.0676, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.8244, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.3422, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.0691, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.4078, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.6943, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.4141, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.5333, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.0336, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.2283, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.6910, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.3113, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.9206, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.9237, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.2552, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.4394, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.8331, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.5119, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.4649, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.0499, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.8802, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.6943, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.8139, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.1715, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.7005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.3744, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.9346, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.4730, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.5391, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.1638, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.2477, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.1332, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.1207, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.2909, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.1532, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.7833, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.2797, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.7360, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.9570, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.1943, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.5823, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.4586, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.4400, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.0257, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.9739, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.0754, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.462406015037594, 'recall': 0.4979757085020243, 'f1': 0.47953216374269003, 'number': 247}, 'P': {'precision': 0.592814371257485, 'recall': 0.72, 'f1': 0.6502463054187192, 'number': 275}, 'overall_precision': 0.535, 'overall_recall': 0.6149425287356322, 'overall_f1': 0.5721925133689839, 'overall_accuracy': 0.7348749272833043}


		-------------RUN 4-----------
			------------EPOCH 1---------------
Loss:  tensor(811.1237, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1299.4048, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1601.7766, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2184.0918, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3071.6880, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2847.6519, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2972.3359, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2044.9268, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1534.6917, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2051.7737, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1495.1387, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2253.6194, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2734.3687, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1379.9373, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1414.9700, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1708.5063, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2237.9614, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1262.5594, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(772.1582, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1362.2189, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1716.9279, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1482.6858, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2100.3804, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2034.5386, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2379.6187, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2339.0198, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1591.3823, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2120.9966, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4044.5640, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1507.1490, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2372.8416, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2357.0713, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2677.8774, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2482.3308, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1819.5876, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1444.4739, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1900.7035, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1316.5210, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(706.3838, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1364.4271, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1448.2976, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1312.3870, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1581.6262, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1401.7480, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1215.6763, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1444.1963, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2316.8943, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(986.0389, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1999.5637, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1429.9045, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2558.3784, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1868.2443, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1663.1433, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1220.2682, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2540.2183, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.03937007874015748, 'recall': 0.04335260115606936, 'f1': 0.04126547455295736, 'number': 346}, 'P': {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'number': 405}, 'overall_precision': 0.031982942430703626, 'overall_recall': 0.019973368841544607, 'overall_f1': 0.024590163934426233, 'overall_accuracy': 0.3641095147736514}
			------------EPOCH 2---------------
Loss:  tensor(511.2300, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(871.5062, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(953.7163, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1285.3860, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1874.9260, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1704.8777, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1974.2065, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1411.5737, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1186.4805, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1451.1731, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1094.5201, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1632.0176, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2058.6299, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(969.7140, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1042.8799, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1382.2415, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1714.4568, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(954.8289, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(578.5594, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1128.3975, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1407.7937, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1127.4120, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1670.7266, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1722.0787, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2049.2046, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1916.5879, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1154.8381, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1690.2949, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3295.3687, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1229.1448, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1916.0242, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2070.5010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2179.7075, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2077.2952, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1438.8319, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1110.6279, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1393.8384, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1126.4403, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(601.9980, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1142.2322, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1218.2709, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1061.1243, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1266.4253, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1187.3256, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(950.4642, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1160.9980, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1864.3502, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(793.8420, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1473.2037, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1077.0396, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1789.3746, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1308.1649, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1204.1492, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1057.4696, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1960.1656, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.12168141592920353, 'recall': 0.15895953757225434, 'f1': 0.13784461152882205, 'number': 346}, 'P': {'precision': 0.15319148936170213, 'recall': 0.08888888888888889, 'f1': 0.11249999999999999, 'number': 405}, 'overall_precision': 0.1324599708879185, 'overall_recall': 0.12117177097203728, 'overall_f1': 0.12656467315716272, 'overall_accuracy': 0.5207373271889401}
			------------EPOCH 3---------------
Loss:  tensor(380.7728, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(796.8405, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(749.9351, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(997.8529, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1239.3423, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1154.7322, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1568.8783, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1092.7251, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(917.6438, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1192.4242, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(861.5736, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1308.0840, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1593.9214, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(757.8366, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(863.5161, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1320.0581, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1458.6267, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(769.0483, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(458.1392, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1033.1965, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1130.0916, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(853.2844, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1436.4700, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1423.4442, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1631.5038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1531.1533, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(877.5271, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1399.5088, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2588.2500, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(932.0520, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1440.0962, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1610.6807, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1650.2266, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1581.0405, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1171.7710, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(902.3149, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1084.0024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(924.5779, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(459.7376, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(890.2693, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(751.9098, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(596.6791, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(799.8863, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(951.2748, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(696.2592, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(909.7146, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1516.7377, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(684.5544, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1086.6650, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(831.3170, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1158.4840, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(941.9388, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(911.3599, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(864.5768, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1522.1863, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.2027729636048527, 'recall': 0.33815028901734107, 'f1': 0.2535211267605634, 'number': 346}, 'P': {'precision': 0.16666666666666666, 'recall': 0.10617283950617284, 'f1': 0.1297134238310709, 'number': 405}, 'overall_precision': 0.19161676646706588, 'overall_recall': 0.21304926764314247, 'overall_f1': 0.20176544766708704, 'overall_accuracy': 0.5545676335050149}
			------------EPOCH 4---------------
Loss:  tensor(312.1058, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(609.6499, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(568.0975, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(786.6336, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(929.7599, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(905.9923, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1232.5527, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(775.8209, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(577.0728, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(876.5934, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(633.4248, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1002.2635, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1253.6072, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(624.6704, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(621.6966, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1029.9875, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(971.0435, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(557.2664, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(321.4306, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(782.5684, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(825.9277, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(668.3234, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1179.7605, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1296.0316, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1164.4064, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1143.3081, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(591.8171, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1033.9617, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2125.6130, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(710.6219, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1021.0511, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1055.7063, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1213.2854, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1193.5554, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(917.1232, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(731.3873, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(854.2357, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(738.6460, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(338.3853, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(679.2152, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(459.0216, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(342.4210, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(458.4200, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(688.0739, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(534.9684, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(711.7897, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1145.0435, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(566.6981, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(881.1939, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(567.6812, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(784.2682, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(756.9081, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(723.2051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(617.5963, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1167.2036, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.2787524366471735, 'recall': 0.41329479768786126, 'f1': 0.33294528521536665, 'number': 346}, 'P': {'precision': 0.17270194986072424, 'recall': 0.15308641975308643, 'f1': 0.162303664921466, 'number': 405}, 'overall_precision': 0.23509174311926606, 'overall_recall': 0.2729693741677763, 'overall_f1': 0.25261860751694387, 'overall_accuracy': 0.6285714285714286}
			------------EPOCH 5---------------
Loss:  tensor(248.2594, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(489.1512, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(355.8862, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(536.3298, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(647.7297, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(646.6099, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(927.5054, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(581.0688, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(380.7822, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(776.9615, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(509.4039, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(812.9910, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(816.4353, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(469.5500, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(445.2438, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(688.9402, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(617.7899, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(324.3042, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(200.9985, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(488.8128, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(522.7671, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(518.4600, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(750.4500, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(923.1910, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(789.8441, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(802.0791, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(401.3636, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(720.1150, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1994.8057, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(605.6625, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(825.7026, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(752.9946, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(888.1342, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(718.3563, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(673.6309, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(507.1336, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(822.2992, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(735.5305, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(293.4022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(637.5383, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(421.1063, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(328.0941, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(386.1171, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(607.7561, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(610.1009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(743.6072, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1174.0640, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(570.0892, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(764.8816, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(576.1304, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(787.1351, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(566.0405, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(585.7775, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(438.7734, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(909.1577, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.35071090047393366, 'recall': 0.4277456647398844, 'f1': 0.38541666666666674, 'number': 346}, 'P': {'precision': 0.19736842105263158, 'recall': 0.25925925925925924, 'f1': 0.22411953041622198, 'number': 405}, 'overall_precision': 0.26519916142557654, 'overall_recall': 0.33688415446071907, 'overall_f1': 0.2967741935483871, 'overall_accuracy': 0.7117375982651125}
			------------EPOCH 6---------------
Loss:  tensor(168.1248, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(419.8519, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(255.4738, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(392.3873, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(423.8171, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(560.8335, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(914.6385, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(533.8685, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(489.1776, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(645.3745, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(473.9797, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(769.1320, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(764.4733, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(418.6814, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(409.0846, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(605.9084, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(508.5020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(282.1254, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(176.7510, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(380.9377, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(457.8948, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(555.6246, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(634.7095, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(684.9019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(672.9860, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(697.3209, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(356.8135, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(523.1505, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1217.6538, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(340.2767, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(532.1981, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(554.2898, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(880.4875, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(730.4053, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(641.1512, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(468.7568, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(681.3230, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(587.4386, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(217.3774, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(474.9386, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(249.8643, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(165.1182, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(218.7987, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(402.8726, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(303.2800, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(456.5583, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(698.8755, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(265.8908, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(454.2278, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(359.3956, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(487.1270, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(372.5244, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(453.0954, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(290.7771, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(808.0329, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.2955665024630542, 'recall': 0.5202312138728323, 'f1': 0.3769633507853404, 'number': 346}, 'P': {'precision': 0.6493506493506493, 'recall': 0.24691358024691357, 'f1': 0.3577817531305903, 'number': 405}, 'overall_precision': 0.3669724770642202, 'overall_recall': 0.37283621837549935, 'overall_f1': 0.36988110964332893, 'overall_accuracy': 0.6132285172133369}
			------------EPOCH 7---------------
Loss:  tensor(166.4348, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(430.9246, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(202.5989, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(419.9906, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(459.6044, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(428.6285, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(660.6367, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(437.3245, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(154.8882, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(450.6512, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(311.0821, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(435.9165, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(465.0298, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(221.1524, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(229.8770, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(432.0650, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(315.0296, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(164.5843, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(143.4907, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(291.4814, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(339.3435, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(427.0989, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(409.7919, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(573.6533, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(362.9595, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(403.7508, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(161.6604, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(282.9506, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1151.2808, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(268.2377, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(365.0303, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(358.6348, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(523.5790, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(428.5645, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(394.1913, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(276.0190, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(409.4366, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(353.3989, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(144.6176, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(340.8610, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(202.6884, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(113.2222, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(133.4011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(279.2996, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(254.8246, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(425.4152, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(607.0674, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(222.9494, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(409.5289, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(284.7629, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(368.9676, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(315.1617, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(508.0268, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(241.4935, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(685.5634, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3195084485407066, 'recall': 0.6011560693641619, 'f1': 0.4172517552657974, 'number': 346}, 'P': {'precision': 0.6850393700787402, 'recall': 0.21481481481481482, 'f1': 0.32706766917293234, 'number': 405}, 'overall_precision': 0.3791773778920309, 'overall_recall': 0.39280958721704395, 'overall_f1': 0.38587311968606935, 'overall_accuracy': 0.5992409867172676}
			------------EPOCH 8---------------
Loss:  tensor(155.0954, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(352.7277, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(159.8782, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(343.0612, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(248.1789, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(302.2694, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(504.7470, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(286.7095, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(116.4025, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(443.3979, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(210.7840, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(387.8002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(276.3279, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(223.7182, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(134.4755, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(270.2904, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(229.0408, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(100.5389, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(77.4740, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(132.2072, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(158.6710, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(163.2456, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(179.3021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(255.7446, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(309.4219, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(330.9832, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(140.7572, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(217.3930, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1340.3254, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(351.8481, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(475.6036, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(397.8003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(723.2286, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(401.5615, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(411.0142, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(415.0645, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(342.9809, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(253.8534, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(173.1567, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(205.7922, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(126.3993, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(95.4186, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(114.1567, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(229.4767, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(149.7040, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(213.4221, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(452.3896, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(141.4935, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(320.2285, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(203.4335, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(239.1151, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(194.1478, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(326.6605, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(179.9682, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(709.6920, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.32220609579100146, 'recall': 0.6416184971098265, 'f1': 0.42898550724637685, 'number': 346}, 'P': {'precision': 0.5054945054945055, 'recall': 0.11358024691358025, 'f1': 0.18548387096774194, 'number': 405}, 'overall_precision': 0.3435897435897436, 'overall_recall': 0.35685752330226367, 'overall_f1': 0.3500979751796212, 'overall_accuracy': 0.5605313092979127}
			------------EPOCH 9---------------
Loss:  tensor(89.6279, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(318.4634, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(119.4811, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(330.6400, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(501.8663, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(441.7461, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(649.0321, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(342.6795, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(82.1557, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(401.2592, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(253.7418, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(484.0670, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(347.3464, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(223.6286, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(196.2920, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(199.1735, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(173.3202, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(82.8430, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(65.0093, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(145.4022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(112.0987, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(176.4161, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(188.7827, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(247.6230, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(200.0770, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(211.4718, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(96.0583, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(133.8525, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(931.4558, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(245.5902, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(255.8633, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(283.3873, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(982.9766, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(820.5443, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(707.1704, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(561.2026, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(602.6068, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(776.6685, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(390.2156, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(576.7654, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(239.1805, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(152.5906, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(290.5836, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(467.0828, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(186.4106, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(206.7506, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(331.8551, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(130.2553, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(270.3963, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(131.4699, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(168.8096, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(160.0065, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(253.4502, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(182.5694, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(349.1330, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.30095541401273884, 'recall': 0.546242774566474, 'f1': 0.38809034907597534, 'number': 346}, 'P': {'precision': 0.8148148148148148, 'recall': 0.10864197530864197, 'f1': 0.19172113289760348, 'number': 405}, 'overall_precision': 0.3416422287390029, 'overall_recall': 0.31025299600532624, 'overall_f1': 0.3251919050942079, 'overall_accuracy': 0.5385741393331526}
			------------EPOCH 10---------------
Loss:  tensor(134.1698, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(425.5847, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(273.0900, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(431.6158, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(666.0220, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(476.7032, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(995.4079, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(426.9957, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(140.2754, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(743.2184, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(392.1396, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(853.0453, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(956.0875, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(713.9746, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(744.6040, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(466.6494, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(718.9112, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(429.3696, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(184.1681, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(516.5323, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(311.2451, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(490.8997, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(630.6063, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(782.1948, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(420.6953, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(407.4529, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(201.3613, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(264.1703, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(610.5092, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(147.8660, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(288.7806, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(306.8803, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(558.5483, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(464.9881, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(352.5438, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(228.7672, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(513.1268, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(680.1371, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(322.1748, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(534.3070, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(543.6931, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(442.9130, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(876.3025, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1196.7235, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(842.0732, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(870.7286, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1188.2307, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(927.4833, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1052.8823, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(813.3926, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(848.0175, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(834.9349, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(513.6646, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(427.2905, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1144.2885, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.41471571906354515, 'recall': 0.3583815028901734, 'f1': 0.38449612403100775, 'number': 346}, 'P': {'precision': 0.47468354430379744, 'recall': 0.5555555555555556, 'f1': 0.5119453924914675, 'number': 405}, 'overall_precision': 0.4514877102199224, 'overall_recall': 0.4647137150466045, 'overall_f1': 0.458005249343832, 'overall_accuracy': 0.7022499322309569}
			------------EPOCH 11---------------
Loss:  tensor(113.5741, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(438.0752, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(207.9114, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(321.2726, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(278.6359, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(438.0907, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(506.1985, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(246.3791, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(256.7729, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(421.8457, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(266.7342, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(506.1224, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(607.8864, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(407.3480, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(361.9146, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(328.1605, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(544.8596, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(319.6417, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(146.5760, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(336.8520, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(570.8557, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(761.8839, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(889.0485, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(654.1396, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(653.8061, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1076.6758, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(877.9094, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1023.6813, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1684.7825, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(490.1452, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(797.1455, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(661.5479, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(999.5389, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(825.5382, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(618.5984, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(440.6424, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(545.9835, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(362.4711, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(163.7390, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(348.5240, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(195.4049, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(141.9289, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(206.7874, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(288.1239, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(300.8345, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(310.1996, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(349.1322, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(229.1743, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(487.1111, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(277.0036, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(314.0261, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(298.3298, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(377.4262, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(300.6847, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(847.6820, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.547945205479452, 'recall': 0.3468208092485549, 'f1': 0.4247787610619469, 'number': 346}, 'P': {'precision': 0.5286738351254481, 'recall': 0.7283950617283951, 'f1': 0.6126687435098651, 'number': 405}, 'overall_precision': 0.5341055341055341, 'overall_recall': 0.5525965379494008, 'overall_f1': 0.543193717277487, 'overall_accuracy': 0.7291406885334779}
			------------EPOCH 12---------------
Loss:  tensor(101.5593, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(488.9721, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(259.9311, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(337.3547, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(220.0594, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(322.7803, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(579.8470, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(390.0358, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(250.4523, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(430.8305, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(399.8218, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(588.4167, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(581.0256, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(279.7765, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(249.2682, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(388.9348, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(270.4676, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(168.9591, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(117.7466, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(391.7583, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(418.2695, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(283.5276, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(222.1235, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(253.8880, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(363.2845, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(455.0206, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(151.5607, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(130.8016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(478.2193, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(95.7482, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(167.4089, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(155.7108, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(527.1442, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(351.8859, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(291.8584, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(159.4501, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(336.2313, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(296.3403, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(153.3004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(342.9473, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(178.4900, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(105.3163, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(119.3849, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(333.6083, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(283.3519, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(532.0706, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1195.6799, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(448.5932, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(308.9974, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(268.3832, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(261.3417, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(212.4183, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(213.3200, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(169.2461, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(390.1211, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4671916010498688, 'recall': 0.5144508670520231, 'f1': 0.4896836313617607, 'number': 346}, 'P': {'precision': 0.5745614035087719, 'recall': 0.6469135802469136, 'f1': 0.6085946573751452, 'number': 405}, 'overall_precision': 0.5256869772998806, 'overall_recall': 0.5858854860186418, 'overall_f1': 0.5541561712846347, 'overall_accuracy': 0.7477907291949037}
			------------EPOCH 13---------------
Loss:  tensor(32.3684, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(154.8158, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(92.9288, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(146.8345, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(110.2859, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(178.5070, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(199.5993, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(72.3462, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(70.8732, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(200.8562, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(119.1302, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(224.8262, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(269.3351, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(148.7554, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(110.2597, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(232.1517, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(135.0839, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(120.1401, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(59.7412, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(127.2843, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(167.1162, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(176.0714, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(172.0781, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(199.1115, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(265.0791, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(304.8460, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(209.3229, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(259.1987, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1466.1910, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(303.3083, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(455.8703, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(259.6830, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(302.1992, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(167.6410, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(142.8138, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(133.4988, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(156.2507, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(134.0829, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.5931, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(121.9528, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(82.0369, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.6662, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(64.5610, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(100.9692, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(116.6261, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(123.0515, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(199.8487, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(87.3369, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(179.2223, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(201.2633, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(194.0169, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(140.7236, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(369.0854, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(474.6447, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(403.2565, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3791469194312796, 'recall': 0.4624277456647399, 'f1': 0.4166666666666667, 'number': 346}, 'P': {'precision': 0.5565217391304348, 'recall': 0.6320987654320988, 'f1': 0.5919075144508671, 'number': 405}, 'overall_precision': 0.47165532879818595, 'overall_recall': 0.5539280958721704, 'overall_f1': 0.509491733006736, 'overall_accuracy': 0.695310382217403}
			------------EPOCH 14---------------
Loss:  tensor(124.6402, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(150.8262, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(672.1863, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(383.5239, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(514.4238, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(558.3282, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(628.0582, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(145.9836, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(101.3082, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(723.6262, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(299.1855, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(422.2836, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(386.9735, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(148.1858, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(248.0577, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(572.2115, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(359.8223, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(103.7233, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(61.2219, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(403.6424, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(234.1805, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(88.6252, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(141.0819, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(151.9539, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(176.6560, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(154.2077, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(72.1434, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(79.7831, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(245.8033, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(75.7944, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(95.6123, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(92.5653, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(295.0791, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(169.0543, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(242.3419, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(127.2210, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(545.2457, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(437.4952, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(89.8316, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(174.2651, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(147.0149, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(124.6416, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(148.2566, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(355.3756, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(325.5773, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(339.6682, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(314.7718, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(432.1719, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(326.8869, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(92.6063, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(218.7321, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(351.8292, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(191.7552, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(173.8366, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(223.5940, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4897360703812317, 'recall': 0.48265895953757226, 'f1': 0.48617176128093154, 'number': 346}, 'P': {'precision': 0.5916473317865429, 'recall': 0.6296296296296297, 'f1': 0.6100478468899523, 'number': 405}, 'overall_precision': 0.5466321243523317, 'overall_recall': 0.5619174434087882, 'overall_f1': 0.5541694024950755, 'overall_accuracy': 0.7472485768500948}
			------------EPOCH 15---------------
Loss:  tensor(16.6773, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(106.3789, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.6762, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(75.3903, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.2895, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(111.4532, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(101.8416, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.4235, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.7393, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(117.1228, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(68.4288, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(103.4957, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(107.1483, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.8205, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.9773, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(84.1197, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(127.4614, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.3126, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.2714, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(133.5385, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(117.4325, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(61.1682, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(76.7431, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(125.4996, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(190.6048, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(240.1150, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(87.0396, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(114.5149, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(216.8699, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(146.1583, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(74.6157, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(131.6952, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(139.2961, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(101.8546, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(97.4959, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(64.7509, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(98.9010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(104.7005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.5739, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(96.1690, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.8628, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.8150, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.1965, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(72.5988, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(73.4058, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(82.1086, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(125.3457, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.6197, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(91.2851, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.4885, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.6332, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(89.6927, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(76.6184, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.9022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(81.8592, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4774011299435028, 'recall': 0.4884393063583815, 'f1': 0.4828571428571428, 'number': 346}, 'P': {'precision': 0.5869565217391305, 'recall': 0.6666666666666666, 'f1': 0.6242774566473989, 'number': 405}, 'overall_precision': 0.5393120393120393, 'overall_recall': 0.5845539280958721, 'overall_f1': 0.5610223642172524, 'overall_accuracy': 0.7336405529953917}
			------------EPOCH 16---------------
Loss:  tensor(18.6728, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.0155, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.8213, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.1144, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.5471, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(85.5056, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(89.0105, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.9691, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.6203, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(78.5375, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.2740, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(71.9840, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(70.4817, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.3663, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.0777, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.7460, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.6406, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.7724, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.9638, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.1725, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.8288, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.5174, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.3932, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(71.9167, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(59.0365, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.4803, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.6492, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.1138, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(107.0789, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.5264, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.7499, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.1727, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(62.6503, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.6258, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.1691, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.6175, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.6484, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.6622, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.2680, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.7261, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.1804, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.4582, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.6723, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.4873, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.6738, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.1438, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(61.8712, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.5674, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(78.7178, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.3646, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.9194, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.7379, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.5510, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.4985, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.6027, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.49855907780979825, 'recall': 0.5, 'f1': 0.49927849927849927, 'number': 346}, 'P': {'precision': 0.6037296037296037, 'recall': 0.6395061728395062, 'f1': 0.6211031175059952, 'number': 405}, 'overall_precision': 0.5567010309278351, 'overall_recall': 0.5752330226364847, 'overall_f1': 0.5658153241650294, 'overall_accuracy': 0.7404174573055028}
			------------EPOCH 17---------------
Loss:  tensor(9.6552, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.3278, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.5677, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.9492, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.4390, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(70.1573, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.9695, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.5951, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.7819, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.7077, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.7143, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.0891, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.2572, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.2780, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.8967, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.2606, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.7627, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.9000, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.3502, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.3845, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.7090, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.2252, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.7421, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.9130, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.9131, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.9041, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.9493, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.0624, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(77.5710, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.1231, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.5458, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.9404, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.3743, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.8366, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.8212, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.0595, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.6902, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.5163, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.1315, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.3616, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.4983, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.0638, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.4227, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.6212, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.3757, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.5740, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.5981, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.7850, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.6067, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.4157, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.5546, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.3143, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.3281, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.1551, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.9640, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5014492753623189, 'recall': 0.5, 'f1': 0.5007235890014472, 'number': 346}, 'P': {'precision': 0.5886363636363636, 'recall': 0.6395061728395062, 'f1': 0.6130177514792899, 'number': 405}, 'overall_precision': 0.5503184713375796, 'overall_recall': 0.5752330226364847, 'overall_f1': 0.5625, 'overall_accuracy': 0.7400921658986175}
			------------EPOCH 18---------------
Loss:  tensor(7.9583, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.0443, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.6523, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.1410, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.1790, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(61.3117, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.1978, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.2928, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.2477, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.7764, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.9223, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.9738, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.9182, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.0040, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.0843, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.5976, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.6697, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.0631, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.5861, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.7791, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.1413, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.8711, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.9484, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.7392, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.9614, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.7165, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.5248, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.4235, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(59.4979, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.3927, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.4448, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.6856, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.5408, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.3763, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.1910, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.7858, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.7553, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.1008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.8813, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.9744, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.3073, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.7750, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.3618, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.9875, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.8301, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.3871, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.6396, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.3088, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.3260, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.7882, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.3877, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.7176, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.7736, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.7957, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.5951, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4899713467048711, 'recall': 0.49421965317919075, 'f1': 0.4920863309352518, 'number': 346}, 'P': {'precision': 0.591324200913242, 'recall': 0.6395061728395062, 'f1': 0.6144721233689205, 'number': 405}, 'overall_precision': 0.5463786531130876, 'overall_recall': 0.5725699067909454, 'overall_f1': 0.5591677503250976, 'overall_accuracy': 0.7373814041745731}
			------------EPOCH 19---------------
Loss:  tensor(7.2169, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.9162, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.8776, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.0952, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.6302, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.6177, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.5507, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.2719, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.9137, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.1218, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.6898, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.9938, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.6405, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.0658, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.9901, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.4979, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.5735, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.2876, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.4366, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.9899, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.4813, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.4189, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.5068, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.9563, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.3095, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.7944, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.1677, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.9301, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.3513, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.7097, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.1288, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.5900, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.5978, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.2358, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.5912, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.8724, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.1904, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.3968, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.2911, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.6903, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.0378, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.1526, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.0130, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.0524, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.1095, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.2456, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.9346, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.7151, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.2334, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.2753, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.6862, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.2619, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.5399, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.4795, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.7517, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.48717948717948717, 'recall': 0.49421965317919075, 'f1': 0.49067431850789095, 'number': 346}, 'P': {'precision': 0.5904977375565611, 'recall': 0.6444444444444445, 'f1': 0.6162927981109799, 'number': 405}, 'overall_precision': 0.544766708701135, 'overall_recall': 0.5752330226364847, 'overall_f1': 0.5595854922279792, 'overall_accuracy': 0.7319598807264841}
			------------EPOCH 20---------------
Loss:  tensor(6.3814, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.8399, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.8464, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.6343, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.2202, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.4887, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.6950, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.0782, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.3516, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.5199, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.7990, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.9192, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.2475, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.8518, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.1593, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.7454, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.5221, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.5666, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.1865, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.2297, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.5700, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.1239, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.2439, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.4234, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.5319, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.4830, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.7456, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.1816, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.1434, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.2763, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.9288, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.9368, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.6838, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.7879, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.6682, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.6468, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.8906, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.9294, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.9135, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.2739, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.3754, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.1232, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.4127, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.4461, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.2713, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.8085, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.3616, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.7861, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.6114, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.8149, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.3469, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.8903, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.9934, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.1223, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.3233, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4943502824858757, 'recall': 0.5057803468208093, 'f1': 0.5, 'number': 346}, 'P': {'precision': 0.5865168539325842, 'recall': 0.6444444444444445, 'f1': 0.6141176470588237, 'number': 405}, 'overall_precision': 0.5456821026282853, 'overall_recall': 0.5805592543275633, 'overall_f1': 0.5625806451612902, 'overall_accuracy': 0.7317430197885606}
			------------EPOCH 21---------------
Loss:  tensor(5.8776, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.2421, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.8830, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.3908, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.3727, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.2096, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.2285, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.1810, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.6417, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.0021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.3398, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.4469, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.6286, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.0903, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.8505, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.7208, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.1104, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.0561, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.9199, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.5975, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.9655, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.5105, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.9122, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.3927, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.2323, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.9380, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.3870, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.8901, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.1719, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.4974, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.1271, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.3731, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.0115, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.5681, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.5405, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.0503, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.0811, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.0068, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.2682, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.9132, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.3757, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.2913, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.2617, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.4004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.7804, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.2318, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.7402, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.7279, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.0908, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.9235, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.7092, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.5339, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.5573, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.8403, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.6501, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5072046109510087, 'recall': 0.5086705202312138, 'f1': 0.5079365079365079, 'number': 346}, 'P': {'precision': 0.5837004405286343, 'recall': 0.654320987654321, 'f1': 0.6169965075669384, 'number': 405}, 'overall_precision': 0.550561797752809, 'overall_recall': 0.5872170439414115, 'overall_f1': 0.5682989690721649, 'overall_accuracy': 0.7343995662781242}
			------------EPOCH 22---------------
Loss:  tensor(5.1376, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.2144, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.7653, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.6217, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.9497, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.1810, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.3844, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.7581, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.9958, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.2606, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.1707, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.6708, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.9948, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.1981, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.3594, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.3508, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.8381, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.3616, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.1381, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.2669, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.6120, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.8357, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.0080, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.6480, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.3610, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.9643, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.8478, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.7279, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.2396, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.2636, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.4462, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.6396, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.7583, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.1032, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.7834, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.4500, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.4007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.7088, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.8750, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.2263, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.2018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.8919, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.9826, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.6975, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.6351, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.8401, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.2509, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.0426, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.0536, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.0742, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.6262, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.8394, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.4857, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.9514, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.0908, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5, 'recall': 0.5144508670520231, 'f1': 0.5071225071225072, 'number': 346}, 'P': {'precision': 0.5898004434589801, 'recall': 0.6567901234567901, 'f1': 0.6214953271028038, 'number': 405}, 'overall_precision': 0.550185873605948, 'overall_recall': 0.5912117177097204, 'overall_f1': 0.569961489088575, 'overall_accuracy': 0.7397668744917322}
			------------EPOCH 23---------------
Loss:  tensor(5.6804, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.0539, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.1777, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.9250, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.2157, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.6991, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.8756, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.5473, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.2627, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.7912, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.5354, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.4786, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.6414, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.4249, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.7720, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.5345, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.2165, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.3095, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.3034, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.5753, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.3314, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.9608, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.7863, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.2352, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.6129, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.6742, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.2891, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.5396, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.5542, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.3127, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.8057, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.5963, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.5618, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.2840, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.2391, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.8797, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.1653, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.2890, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.3608, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.3809, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.4960, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.4643, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.4379, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.7945, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.4338, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.3541, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.9595, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.8565, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.8878, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.1131, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.4500, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.0447, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.3402, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.2655, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.9877, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.47645429362880887, 'recall': 0.49710982658959535, 'f1': 0.48656294200848654, 'number': 346}, 'P': {'precision': 0.5928411633109619, 'recall': 0.654320987654321, 'f1': 0.6220657276995306, 'number': 405}, 'overall_precision': 0.5408415841584159, 'overall_recall': 0.5818908122503329, 'overall_f1': 0.5606157793457345, 'overall_accuracy': 0.7476822987259419}
			------------EPOCH 24---------------
Loss:  tensor(5.2235, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.3310, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.8067, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.5483, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.7801, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.2295, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.4948, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.2252, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.0753, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.1699, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.1742, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.1540, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.1739, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.7802, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.6268, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.0636, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.4813, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.9224, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.8291, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.9067, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.3412, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.9922, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.9486, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.3491, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.8329, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.8917, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.6184, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.6267, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.8537, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.5801, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.3600, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.5310, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.4835, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.0133, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.6878, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.4497, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.4078, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.5336, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.6859, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.9648, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.1906, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.5906, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.3027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.5460, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.0642, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.4037, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.6047, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.4210, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.6499, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.8324, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.8368, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.0075, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.3682, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.7981, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.8723, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.49175824175824173, 'recall': 0.5173410404624278, 'f1': 0.504225352112676, 'number': 346}, 'P': {'precision': 0.5995423340961098, 'recall': 0.6469135802469136, 'f1': 0.6223277909738717, 'number': 405}, 'overall_precision': 0.550561797752809, 'overall_recall': 0.5872170439414115, 'overall_f1': 0.5682989690721649, 'overall_accuracy': 0.7470859311466522}
			------------EPOCH 25---------------
Loss:  tensor(5.1387, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.2359, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.3519, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.5351, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.2169, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.6611, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.9733, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.1636, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.9870, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.3707, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.3952, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.2649, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.4319, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.8967, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.9124, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.6547, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.9563, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.9263, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.5182, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.0375, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.3419, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.4099, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.0635, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.3415, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.0795, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.0214, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.8763, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.7356, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.4381, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.8928, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.0867, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.5225, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.8516, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.3651, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.0038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.4180, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.0240, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.4387, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.9976, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.9650, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.7438, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.0649, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.8534, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.2388, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.0685, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.3483, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.8374, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.7225, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.6615, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.6087, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.2579, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.0736, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.2282, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.9105, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.4422, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.511049723756906, 'recall': 0.5346820809248555, 'f1': 0.5225988700564971, 'number': 346}, 'P': {'precision': 0.6175115207373272, 'recall': 0.6617283950617284, 'f1': 0.6388557806912991, 'number': 405}, 'overall_precision': 0.5690954773869347, 'overall_recall': 0.6031957390146472, 'overall_f1': 0.5856496444731739, 'overall_accuracy': 0.7534833288153971}
			------------EPOCH 26---------------
Loss:  tensor(4.8209, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.1753, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.7211, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.5809, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.5385, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.1035, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.6787, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.2019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.3096, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.1200, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.3012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.3995, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.6432, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.2467, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.5268, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.8017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.7007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.2255, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.6445, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.9171, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.1854, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.9479, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.3990, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.6531, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.5959, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.6004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.5625, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.9935, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.3101, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.6433, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.1050, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.8869, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.6597, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.5093, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.4936, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.8435, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.5029, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.6680, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.3560, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.8745, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.7643, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.7075, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.5680, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.3370, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.4305, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.6698, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.2062, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.3458, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.5196, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.3044, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.3467, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.4351, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.5038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.4668, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.7273, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4823848238482385, 'recall': 0.5144508670520231, 'f1': 0.49790209790209794, 'number': 346}, 'P': {'precision': 0.6152073732718893, 'recall': 0.6592592592592592, 'f1': 0.6364719904648389, 'number': 405}, 'overall_precision': 0.5541718555417185, 'overall_recall': 0.59254327563249, 'overall_f1': 0.5727155727155727, 'overall_accuracy': 0.7518568717809705}
			------------EPOCH 27---------------
Loss:  tensor(4.5883, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.5354, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.7277, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.9665, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.0391, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.9374, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.2781, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.5380, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.7660, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.6031, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.7337, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.2695, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.5374, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.1344, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.9886, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.9629, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.8980, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.0661, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.9822, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.3429, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.3609, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.5077, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.7924, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.0103, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.1132, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.2827, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.0385, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.4398, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.0386, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.0402, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.2235, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.1162, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.7911, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.1767, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.6167, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.0738, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.6757, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.1371, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.0715, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.4680, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.7648, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.8786, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.8022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.1957, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.8801, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.8565, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.0756, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.0263, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.1539, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.0214, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.3201, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.0412, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.9829, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.8798, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.7865, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.48484848484848486, 'recall': 0.5086705202312138, 'f1': 0.49647390691114246, 'number': 346}, 'P': {'precision': 0.6063348416289592, 'recall': 0.6617283950617284, 'f1': 0.6328217237308146, 'number': 405}, 'overall_precision': 0.5515527950310559, 'overall_recall': 0.5912117177097204, 'overall_f1': 0.5706940874035991, 'overall_accuracy': 0.7392247221469233}
			------------EPOCH 28---------------
Loss:  tensor(4.4654, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.2312, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.7704, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.8615, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.8139, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.0372, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.2459, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.1599, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.5003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.6894, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.1625, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.6553, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.5400, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.4469, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.6341, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.4461, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.3452, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.0580, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.5586, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.7822, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.7853, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.0532, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.1158, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.3135, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.5595, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.8753, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.6473, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.9152, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.9099, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.5535, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.6110, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.5606, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.1638, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.6565, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.8604, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.4884, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.8514, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.7047, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.7877, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.0935, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.7196, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.2192, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.1765, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.6658, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.6041, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.4218, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.1994, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.1681, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.4835, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.6162, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.1273, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.8418, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.5193, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.4488, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.0406, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.48753462603878117, 'recall': 0.5086705202312138, 'f1': 0.49787835926449786, 'number': 346}, 'P': {'precision': 0.6099773242630385, 'recall': 0.6641975308641975, 'f1': 0.6359338061465721, 'number': 405}, 'overall_precision': 0.5548628428927681, 'overall_recall': 0.59254327563249, 'overall_f1': 0.5730843528654217, 'overall_accuracy': 0.7381946326917864}
			------------EPOCH 29---------------
Loss:  tensor(4.3099, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.7549, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.3577, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.7872, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.0383, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.6301, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.4130, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.7487, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.1681, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.7953, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.6837, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.9431, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.8030, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.8468, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.1808, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.8715, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.7184, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.1688, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.0919, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.7668, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.6795, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.0861, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.2210, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.8459, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.0962, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.3552, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.2057, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.5060, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.9481, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.1939, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.8211, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.9882, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.4980, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.0130, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.2666, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.0043, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.2590, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.3538, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.4918, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.7101, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.6975, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.4867, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.4743, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.2278, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.1874, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.7060, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.3959, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.4815, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.2698, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.5592, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.1149, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5987, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.1159, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.0241, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.2418, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5, 'recall': 0.5144508670520231, 'f1': 0.5071225071225072, 'number': 346}, 'P': {'precision': 0.6085972850678733, 'recall': 0.6641975308641975, 'f1': 0.6351829988193625, 'number': 405}, 'overall_precision': 0.5601503759398496, 'overall_recall': 0.5952063914780293, 'overall_f1': 0.5771465461588122, 'overall_accuracy': 0.7361344537815127}
			------------EPOCH 30---------------
Loss:  tensor(4.1399, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.3520, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.2140, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.2224, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.4461, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.3881, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.5553, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.4505, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.9394, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.2118, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.1768, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.4432, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.9596, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.3345, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.0114, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.4853, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.3241, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.5573, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.7835, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.2987, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.1010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.9948, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.2833, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.2321, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.6633, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.8322, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.8888, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.1018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.4898, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.7503, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.8472, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.4609, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.7932, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.5881, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.7109, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.6294, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.7178, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.0896, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.2903, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.4283, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.5729, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.7429, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.7722, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.0901, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.9939, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.7868, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.0679, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.0047, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.4719, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.7761, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.4279, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.4459, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.5662, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.6267, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.5741, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4859550561797753, 'recall': 0.5, 'f1': 0.4928774928774929, 'number': 346}, 'P': {'precision': 0.5946547884187082, 'recall': 0.6592592592592592, 'f1': 0.6252927400468384, 'number': 405}, 'overall_precision': 0.546583850931677, 'overall_recall': 0.5858854860186418, 'overall_f1': 0.5655526992287919, 'overall_accuracy': 0.7341284901057197}


		-------------RUN 5-----------
			------------EPOCH 1---------------
Loss:  tensor(2829.6006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3112.8872, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2410.6899, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2312.3164, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1489.1472, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1909.8326, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2277.7441, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2585.7222, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2639.2310, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3399.6265, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3514.4067, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2246.4097, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1129.5499, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1533.9359, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1637.5115, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1106.0891, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2251.8081, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1330.9491, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(921.8052, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2364.3193, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2423.5066, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1664.2764, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1263.5883, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2077.4692, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1451.9106, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1293.5602, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1190.3125, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2301.4995, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1603.7299, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2604.7373, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2246.5073, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2483.2224, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3000.4336, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2390.9285, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2342.4783, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1860.5049, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2046.9543, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1499.6367, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1725.5984, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1591.8577, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1452.4436, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1819.2712, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1756.5231, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2123.2812, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1309.7678, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2448.2007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1504.7280, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(911.6179, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1203.2565, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1033.5696, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1300.1912, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2864.9658, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1381.3977, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2573.1536, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3701.1123, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2250.1704, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2063.6382, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1505.5760, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.18681318681318682, 'recall': 0.08717948717948718, 'f1': 0.11888111888111887, 'number': 195}, 'P': {'precision': 0.12658227848101267, 'recall': 0.18808777429467086, 'f1': 0.15132408575031528, 'number': 319}, 'overall_precision': 0.13628318584070798, 'overall_recall': 0.14980544747081712, 'overall_f1': 0.14272474513438368, 'overall_accuracy': 0.6033302919708029}
			------------EPOCH 2---------------
Loss:  tensor(2068.8911, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2200.2139, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1522.6700, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1527.8590, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1110.7821, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1403.3032, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1729.6687, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1814.0908, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2121.3145, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2745.3203, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2824.8169, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1759.0930, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(878.5229, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1268.4568, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1307.9868, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(839.5642, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1923.4255, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1074.0842, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(735.3278, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1867.3083, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2091.7563, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1331.1472, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(971.3192, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1563.0383, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1218.3860, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1097.4528, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(912.1926, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2208.3682, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1290.0447, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2225.1470, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1817.7441, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2270.0571, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2514.2993, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1899.8628, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1779.6733, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1569.5450, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1675.5906, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1278.9061, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1428.7150, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1282.7981, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1307.2405, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1818.3949, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1707.9641, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1951.4956, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1101.4974, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2046.7227, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1368.5564, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(771.1362, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1041.3285, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(901.0079, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1086.6740, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2246.3142, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1157.9651, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2226.8804, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3339.7908, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1967.2700, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1763.0226, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1338.4390, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.09055118110236221, 'recall': 0.11794871794871795, 'f1': 0.1024498886414254, 'number': 195}, 'P': {'precision': 0.146, 'recall': 0.22884012539184953, 'f1': 0.17826617826617827, 'number': 319}, 'overall_precision': 0.1273209549071618, 'overall_recall': 0.1867704280155642, 'overall_f1': 0.15141955835962145, 'overall_accuracy': 0.6313868613138686}
			------------EPOCH 3---------------
Loss:  tensor(1803.0745, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1865.1083, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1233.6384, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1271.3718, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(954.1039, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1196.2859, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1495.1440, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1513.8958, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1649.2987, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2292.5996, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2400.0210, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1487.9512, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(755.2369, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1114.7810, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1138.5669, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(679.6799, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1676.0210, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(940.5047, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(639.4402, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1432.0022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1627.5847, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1118.3550, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(770.4924, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1264.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1102.5859, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(954.3414, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(753.4110, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1814.2988, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1068.5461, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1793.7841, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1442.3434, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1823.7505, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2091.8062, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1516.6263, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1391.8411, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1308.9766, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1372.8512, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1155.7257, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1257.1177, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1075.2720, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1034.1415, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1481.5049, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1374.3387, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1592.2927, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(851.6414, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1563.1296, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1100.0925, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(622.3086, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(855.6968, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(740.0862, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(929.7230, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1786.4824, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(925.0157, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1865.1074, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2857.4766, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1571.9929, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1457.7584, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1141.1682, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.13768115942028986, 'recall': 0.19487179487179487, 'f1': 0.1613588110403397, 'number': 195}, 'P': {'precision': 0.35626535626535627, 'recall': 0.45454545454545453, 'f1': 0.39944903581267216, 'number': 319}, 'overall_precision': 0.2679355783308931, 'overall_recall': 0.3560311284046693, 'overall_f1': 0.3057644110275689, 'overall_accuracy': 0.6596715328467153}
			------------EPOCH 4---------------
Loss:  tensor(1480.0334, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1550.3105, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(986.5477, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1053.2034, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(772.2906, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(984.4457, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1276.1576, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1233.0896, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1424.7104, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1827.8762, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2027.2458, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1227.6490, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(557.1365, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(824.2849, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(879.0979, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(507.3154, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1319.7198, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(833.0350, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(540.0400, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1122.0583, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1054.9202, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(827.8856, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(539.3466, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1006.0962, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(848.4403, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(673.3441, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(585.9514, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1169.6028, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(765.9184, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1146.9089, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(911.7952, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1324.6237, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1746.7957, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1063.8792, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(998.5458, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(906.2067, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1019.9447, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(938.4393, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(961.9114, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(709.3773, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(643.7609, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(681.6113, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(677.1571, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(927.4360, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(674.3617, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1239.7683, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(837.8157, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(474.2571, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(625.9158, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(500.5959, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(727.4763, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1340.1431, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(660.1600, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1352.9026, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2155.6636, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1015.7991, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1101.4384, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(907.5252, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.2972972972972973, 'recall': 0.4512820512820513, 'f1': 0.35845213849287166, 'number': 195}, 'P': {'precision': 0.5178571428571429, 'recall': 0.6363636363636364, 'f1': 0.5710267229254572, 'number': 319}, 'overall_precision': 0.42296511627906974, 'overall_recall': 0.566147859922179, 'overall_f1': 0.48419301164725453, 'overall_accuracy': 0.739279197080292}
			------------EPOCH 5---------------
Loss:  tensor(1019.5616, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1112.4260, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(693.9537, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(797.1072, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(577.0995, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(708.0670, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(917.0984, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(896.8280, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(765.6218, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1321.6025, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1226.2017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1065.1377, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(476.5846, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(597.3856, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(648.8046, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(471.0792, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(875.6311, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(535.8205, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(420.9197, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(749.9207, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(883.0701, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(651.9563, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(372.1578, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(701.6473, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(691.9036, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(402.0638, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(404.5091, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(778.8845, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(635.7532, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(750.5118, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(659.0168, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(966.2994, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1546.0168, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(817.9496, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(862.1197, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(766.4095, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(686.9043, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(646.0626, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(587.6046, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(502.6901, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(443.9712, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(365.7839, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(375.0254, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(496.1631, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(481.0903, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(892.5443, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(630.4128, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(323.2758, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(401.8104, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(307.6479, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(636.5092, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1055.2102, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(453.0308, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(918.8298, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1914.3616, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(673.4282, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(716.2316, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(661.0364, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3691588785046729, 'recall': 0.40512820512820513, 'f1': 0.38630806845965765, 'number': 195}, 'P': {'precision': 0.47555555555555556, 'recall': 0.670846394984326, 'f1': 0.5565669700910273, 'number': 319}, 'overall_precision': 0.44126506024096385, 'overall_recall': 0.5700389105058365, 'overall_f1': 0.49745331069609505, 'overall_accuracy': 0.697308394160584}
			------------EPOCH 6---------------
Loss:  tensor(864.1328, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1014.4815, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(512.7097, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(705.4867, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(764.8077, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(768.1838, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1081.5176, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(814.1379, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1226.6873, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1801.3831, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1636.2463, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(999.7646, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(316.2591, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(595.2565, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(534.9732, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(346.6675, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(911.2371, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(636.0380, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(451.1156, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(695.6924, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(786.9847, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(631.2932, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(446.0952, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(721.7816, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(711.0468, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(335.0254, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(655.4205, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1311.1489, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(721.4554, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(730.4322, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(787.2445, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(952.9813, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1118.0876, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(603.4745, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(767.8724, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(602.3992, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(622.4462, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(574.5612, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(573.2910, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(465.9623, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(441.8444, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(482.6346, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(536.2524, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(694.9182, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(509.6098, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(925.0264, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(845.4552, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(360.5966, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(394.8353, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(304.6048, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(507.6686, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(918.7911, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(389.9293, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(853.1094, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1371.4001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(468.9286, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(554.3672, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(673.4415, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.2847826086956522, 'recall': 0.6717948717948717, 'f1': 0.4, 'number': 195}, 'P': {'precision': 0.6206896551724138, 'recall': 0.3949843260188088, 'f1': 0.48275862068965525, 'number': 319}, 'overall_precision': 0.38763197586727, 'overall_recall': 0.5, 'overall_f1': 0.4367034834324554, 'overall_accuracy': 0.614279197080292}
			------------EPOCH 7---------------
Loss:  tensor(920.2122, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(908.5087, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(540.6744, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(564.1394, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(384.1721, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(526.1929, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(620.1033, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(743.1061, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(515.1716, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(788.9628, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(685.8125, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(646.5970, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(358.2981, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(521.6466, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(533.6362, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(377.4111, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(634.9473, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(528.0307, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(362.5052, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(546.9899, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(480.9389, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(448.9571, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(296.2208, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(504.9990, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(364.5775, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(194.9727, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(226.7373, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(351.5865, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(381.9308, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(465.8112, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(521.4179, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(653.9422, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(960.4146, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(560.8837, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(755.4378, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(727.9501, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(420.5453, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(393.8331, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(293.1249, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(352.5048, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(192.9111, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(240.1743, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(213.8834, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(250.0988, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(293.5367, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(526.6891, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(420.9367, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(222.1123, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(179.7587, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(145.4801, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(433.1281, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(667.5331, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(301.6241, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(841.4507, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1754.0585, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(407.0983, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(417.7680, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(446.5410, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.36177474402730375, 'recall': 0.5435897435897435, 'f1': 0.4344262295081967, 'number': 195}, 'P': {'precision': 0.5895316804407713, 'recall': 0.670846394984326, 'f1': 0.6275659824046921, 'number': 319}, 'overall_precision': 0.4878048780487805, 'overall_recall': 0.622568093385214, 'overall_f1': 0.5470085470085471, 'overall_accuracy': 0.7016423357664233}
			------------EPOCH 8---------------
Loss:  tensor(261.0749, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(322.3445, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(163.9463, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(212.4429, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(294.5834, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(313.8897, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(408.1273, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(486.8539, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(497.0598, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(856.9877, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(775.3861, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(514.4873, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(266.6130, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(320.9133, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(338.6414, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(282.3410, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(457.7414, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(360.6663, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(301.8477, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(502.4706, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(576.3003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(250.1835, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(132.2020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(252.1412, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(286.8693, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(140.6149, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(159.3799, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(238.3092, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(214.3610, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(272.4807, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(293.4164, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(337.0767, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(449.2357, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(278.3342, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(537.7840, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(190.4666, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(278.2047, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(257.5649, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(288.6196, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(258.3213, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(162.9186, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(225.0996, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(204.9271, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(284.3642, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(393.2734, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(598.1248, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(372.5662, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(208.7040, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(127.5358, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(120.6960, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(280.1959, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(405.7386, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(170.8300, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(460.3340, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(739.5609, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(228.4740, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(277.9539, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(321.1226, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5185185185185185, 'recall': 0.5025641025641026, 'f1': 0.5104166666666666, 'number': 195}, 'P': {'precision': 0.5817307692307693, 'recall': 0.7586206896551724, 'f1': 0.6585034013605442, 'number': 319}, 'overall_precision': 0.5619834710743802, 'overall_recall': 0.6614785992217899, 'overall_f1': 0.6076854334226989, 'overall_accuracy': 0.7598844282238443}
			------------EPOCH 9---------------
Loss:  tensor(209.0853, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(275.6330, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(168.2104, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(231.6578, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(185.3183, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(264.6201, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(361.0743, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(368.4604, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(305.5437, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(427.7725, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(336.5951, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(359.8714, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(177.0133, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(268.0341, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(266.1553, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(270.4683, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(540.0472, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(456.4933, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(325.6611, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(565.7596, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(836.3059, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(513.0623, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(363.0219, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(536.7310, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(426.8751, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(353.8146, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(214.9700, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(350.4184, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(137.7598, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(259.2018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(287.2903, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(334.9248, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(255.5842, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(189.2742, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(303.6991, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(121.7535, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(209.8304, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(228.1850, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(215.2949, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(164.4932, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(124.1629, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(260.1201, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(206.1623, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(215.2251, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(277.6171, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(370.5825, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(308.8640, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(265.7488, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(203.1960, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(151.7855, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(228.0045, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(467.7308, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(227.6057, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(636.3726, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1710.6500, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(477.9525, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(288.0639, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(313.5526, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4748603351955307, 'recall': 0.4358974358974359, 'f1': 0.45454545454545453, 'number': 195}, 'P': {'precision': 0.5728395061728395, 'recall': 0.7272727272727273, 'f1': 0.6408839779005525, 'number': 319}, 'overall_precision': 0.5428082191780822, 'overall_recall': 0.6167315175097277, 'overall_f1': 0.5774134790528233, 'overall_accuracy': 0.7674118004866181}
			------------EPOCH 10---------------
Loss:  tensor(188.0499, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(252.9700, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(117.8739, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(170.5871, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(179.7127, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(170.8572, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(233.3787, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(299.5262, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(310.5941, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(419.9984, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(451.9125, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(420.1357, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(188.6551, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(204.9921, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(285.2634, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(235.2215, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(422.1886, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(303.7003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(231.6636, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(392.1904, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(544.9915, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(288.3013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(111.1779, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(355.6994, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(538.5189, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(370.6342, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(249.8125, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(356.9959, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(378.4731, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(725.2582, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(578.7712, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(841.8300, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(689.8560, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(436.8899, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(450.4382, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(443.9236, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(358.1682, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(173.8689, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(222.4241, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(215.7959, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(115.2266, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(141.2737, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(96.9929, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(133.3428, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(211.0845, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(265.7867, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(200.4781, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(122.8736, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(112.9355, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(90.0674, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(234.4362, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(288.8151, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(158.9700, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(498.1479, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1044.7234, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(238.4560, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(520.6506, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(410.2999, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3181818181818182, 'recall': 0.1794871794871795, 'f1': 0.2295081967213115, 'number': 195}, 'P': {'precision': 0.5563725490196079, 'recall': 0.7115987460815048, 'f1': 0.624484181568088, 'number': 319}, 'overall_precision': 0.5057915057915058, 'overall_recall': 0.5097276264591439, 'overall_f1': 0.507751937984496, 'overall_accuracy': 0.757147201946472}
			------------EPOCH 11---------------
Loss:  tensor(446.5712, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(526.0656, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(251.2032, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(437.0945, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(353.0794, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(157.9300, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(296.5261, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(427.5869, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(225.1748, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(399.2833, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(251.2942, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(422.0506, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(200.8609, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(226.2546, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(247.2174, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(255.8035, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(184.7600, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(146.1199, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(187.4096, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(161.8459, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(116.2887, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(66.8490, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.7442, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(93.5897, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(168.7517, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(61.9119, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(112.5300, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(73.6890, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(105.8743, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(158.6982, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(190.3548, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(313.5499, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(466.9199, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(302.9544, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(291.8802, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(275.1490, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(735.2938, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(381.3714, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(482.9178, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(440.6058, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(239.8035, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(178.6383, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(95.2753, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(283.2124, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(232.2519, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(359.4880, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(278.7588, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(180.6161, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(85.6254, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(84.0456, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(189.2301, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(291.5655, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(100.2953, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(205.0643, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(437.5878, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(100.8678, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(181.1239, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(265.7982, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.49606299212598426, 'recall': 0.3230769230769231, 'f1': 0.391304347826087, 'number': 195}, 'P': {'precision': 0.5605381165919282, 'recall': 0.7836990595611285, 'f1': 0.6535947712418301, 'number': 319}, 'overall_precision': 0.5462478184991274, 'overall_recall': 0.6089494163424124, 'overall_f1': 0.5758969641214351, 'overall_accuracy': 0.7652828467153284}
			------------EPOCH 12---------------
Loss:  tensor(207.0428, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(273.6172, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(86.8941, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(129.6420, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(252.9697, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(172.3429, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(254.5471, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(309.1466, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(129.0038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(275.0476, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(219.2067, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(240.4252, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(110.5070, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(125.3089, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(179.1192, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(197.2276, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(183.8611, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(127.7380, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(154.7652, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(176.3117, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(208.0722, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(88.3543, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(81.0434, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(142.3620, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(161.2458, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(70.0277, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(151.8476, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(103.4439, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(94.7385, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(124.5922, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(235.9386, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(208.9491, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(178.3012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(130.4922, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(148.1191, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(94.8444, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(126.7315, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(114.2065, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(105.5210, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(66.9316, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(71.8652, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(152.8055, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(85.5023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(127.2277, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(150.3117, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(401.7756, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(147.7596, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(138.3538, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(107.4251, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(91.5725, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(403.3051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(628.3624, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(67.4601, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(526.9977, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(366.7158, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(89.4273, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(102.7562, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(172.2431, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5077720207253886, 'recall': 0.5025641025641026, 'f1': 0.5051546391752577, 'number': 195}, 'P': {'precision': 0.5825, 'recall': 0.7304075235109718, 'f1': 0.6481223922114048, 'number': 319}, 'overall_precision': 0.5581787521079258, 'overall_recall': 0.6439688715953308, 'overall_f1': 0.5980126467931345, 'overall_accuracy': 0.7731903892944039}
			------------EPOCH 13---------------
Loss:  tensor(55.3265, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(83.7922, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.5715, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(65.0136, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(90.5101, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.5759, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(102.6231, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(147.0260, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(221.1471, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(284.7915, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(316.8513, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(347.7487, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(239.1356, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(304.4251, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(585.1243, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(301.6601, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(342.7415, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(233.1113, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(212.4891, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(304.3145, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(155.6417, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(93.6657, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(110.6983, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(107.9572, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(97.6132, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.7797, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(82.8395, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.7646, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.0562, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(89.8097, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(134.7354, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(128.8739, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(82.0209, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.4941, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(78.4095, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.9864, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(97.1181, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(65.9396, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(65.6726, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(70.3620, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(76.4179, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(66.5535, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.7667, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.7571, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(179.6191, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(241.2685, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(242.3847, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(168.2087, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(163.2927, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(123.3873, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(272.2534, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(580.1219, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(277.9757, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1013.6498, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1012.1593, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(391.1830, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(218.9083, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(521.5894, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4166666666666667, 'recall': 0.6666666666666666, 'f1': 0.5128205128205129, 'number': 195}, 'P': {'precision': 0.6468531468531469, 'recall': 0.5799373040752351, 'f1': 0.6115702479338844, 'number': 319}, 'overall_precision': 0.5267558528428093, 'overall_recall': 0.6128404669260701, 'overall_f1': 0.5665467625899281, 'overall_accuracy': 0.7401155717761557}
			------------EPOCH 14---------------
Loss:  tensor(224.1522, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(247.3953, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(62.3070, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(73.8996, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(62.1844, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.6104, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(93.4034, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(131.0081, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(96.0531, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(108.3835, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(117.6871, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(138.0904, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(75.2024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(82.9533, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(129.3825, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(94.8655, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(413.8516, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(343.6160, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(244.5327, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(418.0134, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(681.9138, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(888.9080, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(434.7818, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(994.5416, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(278.8922, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(848.0037, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(441.1505, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1304.5779, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(537.3571, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(358.3973, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(420.2173, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(432.6903, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(230.4099, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(82.6141, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(328.6391, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(154.5228, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(78.3005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.2061, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(61.0456, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.3638, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.9375, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(59.3072, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.9990, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.8975, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(111.4163, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(132.7847, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(89.7095, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(73.8031, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(185.3199, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(153.9088, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(235.8940, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(574.2065, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(228.9142, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(700.6036, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1250.9800, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(856.9398, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(555.0350, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(505.8019, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3274336283185841, 'recall': 0.7589743589743589, 'f1': 0.4574961360123647, 'number': 195}, 'P': {'precision': 0.6619718309859155, 'recall': 0.14733542319749215, 'f1': 0.24102564102564103, 'number': 319}, 'overall_precision': 0.372848948374761, 'overall_recall': 0.37937743190661477, 'overall_f1': 0.3760848601735776, 'overall_accuracy': 0.5565693430656934}
			------------EPOCH 15---------------
Loss:  tensor(975.2715, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(934.8189, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(355.8641, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(510.1580, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(366.1721, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(593.1548, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(642.4741, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(670.7119, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(411.3399, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1364.4094, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(662.1127, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(206.0940, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(157.8192, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(134.3215, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(164.7002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(133.5307, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(158.9386, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(145.0388, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(105.6551, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(157.7863, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(198.0760, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(208.9326, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(202.6479, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(381.0048, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(332.4095, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(266.8554, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(353.8900, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(285.6505, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(399.6141, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(405.1110, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(461.0509, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(722.0338, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1518.1091, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(717.7123, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(734.8859, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(621.6862, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(393.3442, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(356.8855, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(346.9311, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(216.7155, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(188.5184, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(271.3443, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(280.9771, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(306.5702, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(199.2295, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(317.8364, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(184.0807, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(91.2050, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(76.8139, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(66.9039, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(141.4086, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(256.9344, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(85.3253, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(206.1205, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(425.2552, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(170.7649, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(182.6444, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(143.2888, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.37228260869565216, 'recall': 0.7025641025641025, 'f1': 0.4866785079928952, 'number': 195}, 'P': {'precision': 0.5971563981042654, 'recall': 0.3949843260188088, 'f1': 0.4754716981132075, 'number': 319}, 'overall_precision': 0.4542314335060449, 'overall_recall': 0.5116731517509727, 'overall_f1': 0.48124428179322964, 'overall_accuracy': 0.6745742092457421}
			------------EPOCH 16---------------
Loss:  tensor(157.6623, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(209.5286, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(97.0523, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(124.6485, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(175.2482, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(274.8376, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(252.2549, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(290.1154, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(201.7798, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(711.9413, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(647.8998, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(252.7789, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(367.2545, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(269.8508, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(225.4604, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(233.6992, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(386.1136, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(251.7452, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(178.9509, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(235.4859, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(251.8665, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(172.4973, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(102.7355, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(231.7809, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(247.4984, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(62.6105, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.8457, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(149.4468, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(141.1888, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(155.4184, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(226.6505, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(161.9299, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(160.8656, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(90.7483, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(99.3788, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(76.9673, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(122.2326, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(76.2246, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(81.4911, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(62.0421, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.4489, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(89.3273, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(85.3479, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(98.0777, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(121.5249, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(149.7780, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.5877, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(77.3459, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(65.1034, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(64.2158, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(110.7302, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(218.8582, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(67.7539, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(294.6561, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(326.1274, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(83.5397, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(198.8835, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(125.0527, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.518918918918919, 'recall': 0.49230769230769234, 'f1': 0.5052631578947369, 'number': 195}, 'P': {'precision': 0.6237373737373737, 'recall': 0.774294670846395, 'f1': 0.6909090909090908, 'number': 319}, 'overall_precision': 0.5903614457831325, 'overall_recall': 0.6673151750972762, 'overall_f1': 0.6264840182648401, 'overall_accuracy': 0.7877889294403893}
			------------EPOCH 17---------------
Loss:  tensor(72.7037, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(101.9000, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.1712, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(80.9632, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(82.6464, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(95.6659, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(141.5533, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(205.6528, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(86.1050, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(133.8828, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(171.5648, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(138.1317, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(72.0199, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(86.1393, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(100.2062, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.2993, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(75.6702, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.8862, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(68.0071, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(80.0944, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(71.6940, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.4125, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.4402, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(72.0026, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(75.0474, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.7768, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.2061, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.7382, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.1910, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.5980, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(83.9827, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(91.4013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(91.5728, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.6572, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(65.3012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.1329, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.1660, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.1561, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.9143, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.1423, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.0567, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.6850, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.9901, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.2657, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(75.0309, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(72.0454, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.8101, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.0927, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.1540, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.3076, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(76.3961, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(133.4158, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.8647, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(75.8490, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(163.5128, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.9678, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(59.0545, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.0939, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.41605839416058393, 'recall': 0.5846153846153846, 'f1': 0.48614072494669514, 'number': 195}, 'P': {'precision': 0.6030769230769231, 'recall': 0.6144200626959248, 'f1': 0.608695652173913, 'number': 319}, 'overall_precision': 0.5175292153589316, 'overall_recall': 0.603112840466926, 'overall_f1': 0.5570530098831985, 'overall_accuracy': 0.746198296836983}
			------------EPOCH 18---------------
Loss:  tensor(33.4772, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.8642, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.9242, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.9084, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.1550, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.4292, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(75.0945, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(90.3845, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.5725, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(79.3281, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(78.9508, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.0294, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(62.6244, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.2056, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(61.5743, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.9285, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.4099, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.5269, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.1387, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.0065, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.2743, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.6492, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.0363, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.1580, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.8341, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.2911, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.3564, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.8524, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.7990, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.1735, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(68.2742, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.5052, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(72.0604, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.5998, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.6897, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.4860, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.0755, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.0379, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.9065, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.8445, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.8886, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.2563, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.2858, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.7052, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(64.3748, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.5068, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.8210, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.1461, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.5673, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.6837, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.3124, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(119.8329, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.7965, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(61.2926, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(146.6830, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.8554, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.0390, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.5982, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4423076923076923, 'recall': 0.5897435897435898, 'f1': 0.5054945054945056, 'number': 195}, 'P': {'precision': 0.6172839506172839, 'recall': 0.6269592476489029, 'f1': 0.6220839813374806, 'number': 319}, 'overall_precision': 0.5393835616438356, 'overall_recall': 0.6128404669260701, 'overall_f1': 0.5737704918032787, 'overall_accuracy': 0.745742092457421}
			------------EPOCH 19---------------
Loss:  tensor(33.3503, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.0112, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.0885, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.8909, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.9356, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.1808, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.2567, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(62.2666, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.7130, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(71.7996, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(80.0116, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.5950, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.7412, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.9935, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.7681, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.3235, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.0705, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.4634, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.7014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.9180, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.6672, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.3129, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.1779, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.6835, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.0257, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.9471, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.0640, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.7464, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.7388, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.8111, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(61.7023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.2405, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.8801, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.9930, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.6904, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.6620, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.0237, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.3940, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.2017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.0913, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.8821, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.9692, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.0435, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.7967, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.3130, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.1929, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.9441, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.3752, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.1772, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.7344, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.4057, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(108.2695, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.7552, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.7846, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(123.7476, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.6285, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.0749, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.1252, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4470588235294118, 'recall': 0.5846153846153846, 'f1': 0.5066666666666666, 'number': 195}, 'P': {'precision': 0.6012084592145015, 'recall': 0.6238244514106583, 'f1': 0.6123076923076923, 'number': 319}, 'overall_precision': 0.5341296928327645, 'overall_recall': 0.6089494163424124, 'overall_f1': 0.569090909090909, 'overall_accuracy': 0.7450577858880778}
			------------EPOCH 20---------------
Loss:  tensor(33.0071, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.7314, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.7708, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.7051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.2735, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.1107, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.9936, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.2753, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(64.0383, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.3217, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.7415, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.4850, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.2609, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.5679, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.9106, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.6015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.0300, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.9500, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.2636, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.6104, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.4636, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.7503, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.0856, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.4245, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.7548, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.4974, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.9807, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.2002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.7465, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.9719, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.6765, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.8475, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.4608, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.6853, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.0855, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.4704, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.7709, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.5282, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.9847, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.4109, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.0649, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.5634, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.4689, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.5493, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.2828, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.8075, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.4044, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.8770, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.7826, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.3024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(93.4554, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.9755, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.6622, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(86.3785, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.9139, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.6146, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.6509, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.43529411764705883, 'recall': 0.5692307692307692, 'f1': 0.49333333333333335, 'number': 195}, 'P': {'precision': 0.6193353474320241, 'recall': 0.6426332288401254, 'f1': 0.6307692307692307, 'number': 319}, 'overall_precision': 0.5392491467576792, 'overall_recall': 0.6147859922178989, 'overall_f1': 0.5745454545454546, 'overall_accuracy': 0.7471867396593674}
			------------EPOCH 21---------------
Loss:  tensor(30.0988, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.0558, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.3899, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.2786, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.4805, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.6533, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.2620, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.8926, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.9375, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.9644, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.9692, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.0935, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.1831, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.0054, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.7579, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.6111, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.7199, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.4174, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.2923, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.6998, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.8885, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.1654, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.0970, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.1599, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.0603, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.8204, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.6240, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.5492, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.2473, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.8455, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.9056, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.3119, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.7298, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.7679, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.7644, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.9436, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.8272, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.5128, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.9165, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.7042, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.8710, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.0906, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.6684, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.2649, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.9837, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.2961, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.5123, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.8106, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.5878, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.3719, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.1052, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(82.3805, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.8959, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.1016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(68.8507, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.5124, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.9855, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.7162, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.44841269841269843, 'recall': 0.5794871794871795, 'f1': 0.505592841163311, 'number': 195}, 'P': {'precision': 0.6096096096096096, 'recall': 0.6363636363636364, 'f1': 0.6226993865030674, 'number': 319}, 'overall_precision': 0.5401709401709401, 'overall_recall': 0.6147859922178989, 'overall_f1': 0.5750682438580528, 'overall_accuracy': 0.7479470802919708}
			------------EPOCH 22---------------
Loss:  tensor(24.7234, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.4639, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.4789, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.0365, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.1850, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.8307, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.6489, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.5834, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.8732, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.6113, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.8300, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.8353, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.7999, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.6579, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.1080, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.9896, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.9017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.7819, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.7460, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.7705, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.8993, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.7750, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.5214, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.2133, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.0340, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.0708, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.2607, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.4031, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.3957, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.7305, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.3277, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.7068, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.7849, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.2698, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.3571, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.1657, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.0306, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.1868, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.3896, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.6337, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.4586, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.6248, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.2060, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.4664, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.9921, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.8519, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.4514, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.0799, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.8366, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.2289, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.0743, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(77.7679, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.6992, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.7480, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.3941, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.4712, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.0286, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.8035, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.44047619047619047, 'recall': 0.5692307692307692, 'f1': 0.4966442953020134, 'number': 195}, 'P': {'precision': 0.6107784431137725, 'recall': 0.6394984326018809, 'f1': 0.6248085758039817, 'number': 319}, 'overall_precision': 0.537542662116041, 'overall_recall': 0.6128404669260701, 'overall_f1': 0.5727272727272729, 'overall_accuracy': 0.7486313868613139}
			------------EPOCH 23---------------
Loss:  tensor(23.0844, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.3182, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.1682, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.5318, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.6368, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.1793, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.5847, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.6131, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.8672, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.4370, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.6574, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.1606, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.3069, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.2456, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.9423, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.9088, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.2594, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.3750, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.2200, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.5362, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.6295, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.8508, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.7966, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.4435, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.7749, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.5931, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.8982, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.6231, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.9939, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.7247, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.6963, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.8625, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.8447, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.8019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.5264, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.4063, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.9399, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.8905, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.6192, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.8567, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.9245, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.9797, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.6577, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.0621, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.5933, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.2650, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.4569, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.7564, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.4452, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.0918, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.1072, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(75.0935, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.7416, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.8700, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.4268, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.0755, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.7523, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.2227, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.44621513944223107, 'recall': 0.5743589743589743, 'f1': 0.5022421524663677, 'number': 195}, 'P': {'precision': 0.6102719033232629, 'recall': 0.6332288401253918, 'f1': 0.6215384615384615, 'number': 319}, 'overall_precision': 0.5395189003436426, 'overall_recall': 0.6108949416342413, 'overall_f1': 0.572992700729927, 'overall_accuracy': 0.7455900243309003}
			------------EPOCH 24---------------
Loss:  tensor(23.0119, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.8182, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.2361, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.8208, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.4450, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.6250, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.6452, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.6410, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.0908, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.6934, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.8129, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.2437, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.7149, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.0031, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.8600, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.2394, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.7272, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.6350, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.8300, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.8859, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.6480, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.4664, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.6910, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.7362, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.2577, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.5687, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.3356, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.3718, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.5147, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.6051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.1609, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.1389, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.1118, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.8592, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.9181, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.8387, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.2207, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.1142, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.7418, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.6702, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.7449, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.6052, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.2049, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.6164, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.6154, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.0033, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.7263, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.1833, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.3650, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.2513, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.9083, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(73.1470, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.0680, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.2158, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.8518, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.3175, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.3316, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.8248, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4314516129032258, 'recall': 0.5487179487179488, 'f1': 0.4830699774266366, 'number': 195}, 'P': {'precision': 0.6201780415430267, 'recall': 0.6551724137931034, 'f1': 0.6371951219512195, 'number': 319}, 'overall_precision': 0.5401709401709401, 'overall_recall': 0.6147859922178989, 'overall_f1': 0.5750682438580528, 'overall_accuracy': 0.7476429440389294}
			------------EPOCH 25---------------
Loss:  tensor(26.8859, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.0149, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.1603, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.4041, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.3385, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.3112, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.2899, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.5711, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.5890, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.5520, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.4468, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.2432, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.6400, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.7773, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.2827, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.4647, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.8452, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.1522, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.9347, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.9203, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.4891, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.4880, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.8173, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.7724, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.7817, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.9912, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.7695, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.2483, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.7420, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.4260, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.4663, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.6438, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.7558, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.8103, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.1276, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.0724, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.3246, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.3643, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.4704, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.0557, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.0158, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.5833, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.6253, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.4168, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.6311, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.1604, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.1396, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.1088, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.0662, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.8623, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.2693, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(70.5277, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.5087, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.8097, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.8471, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.7913, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.8084, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.1724, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.43902439024390244, 'recall': 0.5538461538461539, 'f1': 0.489795918367347, 'number': 195}, 'P': {'precision': 0.6283987915407855, 'recall': 0.6520376175548589, 'f1': 0.6399999999999999, 'number': 319}, 'overall_precision': 0.5476603119584056, 'overall_recall': 0.6147859922178989, 'overall_f1': 0.5792850595783685, 'overall_accuracy': 0.7455900243309003}
			------------EPOCH 26---------------
Loss:  tensor(25.1661, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.6963, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.9540, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.2817, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.6209, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.8554, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.5281, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.0228, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.7267, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.1043, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.2733, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.4195, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.6165, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.7496, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.1476, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.7304, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.5009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.6492, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.0945, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.5459, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.0593, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.4052, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.2571, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.9870, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.3506, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.2878, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.2804, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.9080, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.7781, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.9060, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.9465, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.6627, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.3790, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.0550, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.2442, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.5913, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.5737, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.2257, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.4921, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.5280, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.5624, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.1514, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.1816, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.2069, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.6829, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.2737, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.7111, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.3350, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.1633, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.3176, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.0083, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(65.9498, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.8066, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.3189, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.1187, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.8983, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.7567, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.3722, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.42063492063492064, 'recall': 0.5435897435897435, 'f1': 0.4742729306487696, 'number': 195}, 'P': {'precision': 0.6156156156156156, 'recall': 0.6426332288401254, 'f1': 0.628834355828221, 'number': 319}, 'overall_precision': 0.5316239316239316, 'overall_recall': 0.6050583657587548, 'overall_f1': 0.5659690627843493, 'overall_accuracy': 0.7411800486618005}
			------------EPOCH 27---------------
Loss:  tensor(20.6599, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.7324, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.2857, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.3738, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.5281, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.9213, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.6492, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.9814, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.2471, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.1744, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.6207, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.7642, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.4214, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.7786, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.8625, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.2141, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.7334, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.2802, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.6638, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.0254, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.2040, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.4356, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.3121, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.9551, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.8557, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.9270, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.2690, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.9124, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.2124, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.0441, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.0344, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.3639, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.4859, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.5398, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.7376, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.6885, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.4316, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.1041, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.9546, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.5490, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.9090, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.1457, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.1434, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.2333, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.6673, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.8046, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.1445, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.2558, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.0757, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.6158, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.3353, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(59.9150, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.6544, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.3766, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.3447, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.2013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.0667, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.4343, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4, 'recall': 0.5641025641025641, 'f1': 0.46808510638297873, 'number': 195}, 'P': {'precision': 0.5976331360946746, 'recall': 0.6332288401253918, 'f1': 0.6149162861491628, 'number': 319}, 'overall_precision': 0.5089722675367048, 'overall_recall': 0.6070038910505836, 'overall_f1': 0.5536823425022181, 'overall_accuracy': 0.7340328467153284}
			------------EPOCH 28---------------
Loss:  tensor(23.5112, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.7987, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.6419, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.1632, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.2879, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.9326, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.3118, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.8432, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.9357, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.4528, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.2309, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.2460, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.1627, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.5038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.8055, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.4130, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.3683, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.6479, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.6681, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.1817, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.2551, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.7404, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.1374, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.4564, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.7180, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.8269, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.4947, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.0461, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.1375, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.3568, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.8192, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.9150, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.6613, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.4794, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.7756, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.6793, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.4420, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.0941, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.1110, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.0241, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.8220, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.4722, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.0975, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.7169, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.7170, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.1768, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.2329, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.2933, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.8608, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.2616, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.7147, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.7802, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.6192, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.0950, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.4510, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.2119, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.4212, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.6268, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.43724696356275305, 'recall': 0.5538461538461539, 'f1': 0.4886877828054299, 'number': 195}, 'P': {'precision': 0.6200607902735562, 'recall': 0.6394984326018809, 'f1': 0.6296296296296297, 'number': 319}, 'overall_precision': 0.5416666666666666, 'overall_recall': 0.6070038910505836, 'overall_f1': 0.5724770642201834, 'overall_accuracy': 0.7379105839416058}
			------------EPOCH 29---------------
Loss:  tensor(18.0863, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.1201, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.8152, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.4396, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.8735, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.9026, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.6828, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.2690, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.0183, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.5681, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.3207, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.6917, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.5250, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.9645, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.9633, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.9340, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.7743, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.6567, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.0392, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.6923, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.6945, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.4428, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.2024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.6964, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.9232, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.0385, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.1884, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.9556, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.7252, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.2105, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.1444, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.0989, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.5235, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.8462, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.2862, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.5441, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.2963, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.0934, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.2569, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.2089, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.7265, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.6386, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.9248, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.8303, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.2952, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.2825, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.9215, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.1085, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.2599, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.5177, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.0388, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.9893, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.7638, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.8627, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.0632, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.7963, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.3391, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.3304, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4457831325301205, 'recall': 0.5692307692307692, 'f1': 0.5, 'number': 195}, 'P': {'precision': 0.6242424242424243, 'recall': 0.64576802507837, 'f1': 0.6348228043143298, 'number': 319}, 'overall_precision': 0.5474956822107081, 'overall_recall': 0.6167315175097277, 'overall_f1': 0.5800548947849954, 'overall_accuracy': 0.7395833333333334}
			------------EPOCH 30---------------
Loss:  tensor(18.0390, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.2762, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.9743, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.7906, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.2382, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.1459, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.7688, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.9182, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.6127, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.2300, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.7845, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.7310, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.0857, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.4205, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.4168, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.4617, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.4180, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.2687, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.1940, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.7849, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.7262, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.9683, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.8050, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.8655, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.5982, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.1896, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.8036, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.2004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.6422, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.7101, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.7571, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.0890, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.9492, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.6223, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.8441, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.6435, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.7613, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.8178, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.1159, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.5076, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.5781, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.8469, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.6479, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.5233, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.1711, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.7976, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.7471, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.6823, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.8944, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.2312, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.3517, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.7552, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.8848, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.0914, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.8992, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.5874, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.0195, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.2873, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.45528455284552843, 'recall': 0.5743589743589743, 'f1': 0.5079365079365078, 'number': 195}, 'P': {'precision': 0.6156156156156156, 'recall': 0.6426332288401254, 'f1': 0.628834355828221, 'number': 319}, 'overall_precision': 0.5474956822107081, 'overall_recall': 0.6167315175097277, 'overall_f1': 0.5800548947849954, 'overall_accuracy': 0.7414081508515815}
	Train size: 50 Test size: 50


		-------------RUN 1-----------
			------------EPOCH 1---------------
Loss:  tensor(2530.0625, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2552.9824, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3591.3120, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1587.9639, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2503.0295, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2776.4219, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3614.8916, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3121.2961, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2962.6226, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2575.0364, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2051.8743, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1912.0498, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1568.3325, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(659.2657, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1107.5227, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1839.4698, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2656.0879, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2500.2253, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2533.9495, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1919.1132, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1986.2573, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1783.7317, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1706.2527, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2030.7180, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1499.8328, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2147.9023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2345.8652, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1793.1825, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1228.9023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2121.7832, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1692.1965, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2703.7847, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2918.0940, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.013513513513513514, 'recall': 0.007471980074719801, 'f1': 0.009623095429029671, 'number': 803}, 'P': {'precision': 0.00392156862745098, 'recall': 0.0012106537530266344, 'f1': 0.0018501387604070304, 'number': 826}, 'overall_precision': 0.010014306151645207, 'overall_recall': 0.004297114794352364, 'overall_f1': 0.006013745704467354, 'overall_accuracy': 0.32603317811408616}
			------------EPOCH 2---------------
Loss:  tensor(2055.9988, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1748.0619, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2373.6235, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1216.9208, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2232.1309, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2271.0811, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3052.7549, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2503.1780, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2404.8223, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2097.2727, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1774.9640, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1565.4692, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1303.8159, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(584.4444, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(974.5314, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1480.7734, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2317.2983, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2075.0386, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2115.7422, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1632.8832, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1705.6598, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1548.2087, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1527.7870, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1811.9736, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1338.8994, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1895.2762, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2091.0352, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1574.6283, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1147.8551, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2078.4575, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1695.1804, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2754.7312, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2665.2480, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.09574468085106383, 'recall': 0.033623910336239106, 'f1': 0.049769585253456226, 'number': 803}, 'P': {'precision': 0.25804529201430276, 'recall': 0.5242130750605327, 'f1': 0.34584664536741216, 'number': 826}, 'overall_precision': 0.23469387755102042, 'overall_recall': 0.2823818293431553, 'overall_f1': 0.25633881303984396, 'overall_accuracy': 0.538465269693442}
			------------EPOCH 3---------------
Loss:  tensor(1700.7334, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1735.3865, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2437.3804, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(997.9186, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1677.4795, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2052.7258, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2601.3726, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2011.1362, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1910.7163, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1827.5804, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1390.4092, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1394.9331, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1180.5579, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(465.5322, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(856.8521, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1233.8848, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2026.7231, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1921.4868, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1963.1519, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1498.3147, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1423.3767, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1350.4674, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1336.3245, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1664.7173, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1118.2500, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1543.0896, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1773.2668, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1355.2034, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(951.9769, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1731.7188, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1483.8899, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2468.1145, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2341.3967, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.09722222222222222, 'recall': 0.10460772104607721, 'f1': 0.10077984403119375, 'number': 803}, 'P': {'precision': 0.1459346768589298, 'recall': 0.2542372881355932, 'f1': 0.18543046357615892, 'number': 826}, 'overall_precision': 0.1276595744680851, 'overall_recall': 0.18047882136279927, 'overall_f1': 0.14954221770091555, 'overall_accuracy': 0.5903181994567326}
			------------EPOCH 4---------------
Loss:  tensor(1476.4131, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1432.3336, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2065.3059, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(824.2509, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1502.2041, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1822.0603, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2213.7280, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1650.5883, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1611.3469, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1590.9182, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1170.3435, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1161.1216, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(987.4907, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(388.9249, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(783.9095, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(999.5902, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1710.0090, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1624.2878, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1742.5496, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1305.6985, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1182.0237, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1171.5717, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1143.3005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1459.6108, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(911.9824, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1246.1871, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1414.6718, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1120.2117, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(789.1334, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1412.8679, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1317.6584, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2135.4438, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1972.8733, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.09684361549497848, 'recall': 0.16811955168119552, 'f1': 0.12289485662266728, 'number': 803}, 'P': {'precision': 0.23233532934131737, 'recall': 0.23486682808716708, 'f1': 0.23359422034918723, 'number': 826}, 'overall_precision': 0.14759982054733065, 'overall_recall': 0.2019643953345611, 'overall_f1': 0.17055469155002592, 'overall_accuracy': 0.591361078773768}
			------------EPOCH 5---------------
Loss:  tensor(1299.9265, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1164.3613, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1731.1047, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(725.7059, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1373.8408, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1642.9434, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1908.1245, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1432.0032, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1404.2927, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1408.0859, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(980.7712, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1009.3168, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(814.9590, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(317.5630, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(745.1549, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(697.4431, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1299.7327, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1389.6277, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1536.4148, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1075.2184, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1092.3846, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1084.2504, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(996.4245, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1317.6482, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(753.9755, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1053.0593, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1099.8352, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(804.9991, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(588.8562, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1002.5958, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1126.7207, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1804.2573, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1613.0731, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.1298904538341158, 'recall': 0.31008717310087175, 'f1': 0.18308823529411763, 'number': 803}, 'P': {'precision': 0.326007326007326, 'recall': 0.21549636803874092, 'f1': 0.25947521865889217, 'number': 826}, 'overall_precision': 0.17336581404790904, 'overall_recall': 0.26212400245549416, 'overall_f1': 0.20869990224828933, 'overall_accuracy': 0.5825329840900272}
			------------EPOCH 6---------------
Loss:  tensor(1061.5933, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(943.1069, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1506.8726, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(636.8192, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1148.0999, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1412.8700, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1607.7919, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1215.3090, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1173.2378, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1199.1367, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(759.0674, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(838.1462, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(667.7289, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(248.1422, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(661.4290, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(511.0964, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(980.4049, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1110.1089, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1259.9247, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(826.3750, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(989.3547, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(917.8074, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(857.4603, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1180.4866, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(566.8987, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(797.5872, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(796.5405, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(524.2280, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(476.0973, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(677.9600, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(964.7803, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1477.5592, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1307.0446, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.1878999418266434, 'recall': 0.40224159402241594, 'f1': 0.2561459159397304, 'number': 803}, 'P': {'precision': 0.4126126126126126, 'recall': 0.2772397094430993, 'f1': 0.331643736422882, 'number': 826}, 'overall_precision': 0.24274406332453827, 'overall_recall': 0.3388581952117864, 'overall_f1': 0.2828593389700231, 'overall_accuracy': 0.5886689949553745}
			------------EPOCH 7---------------
Loss:  tensor(923.6545, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(799.6415, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1369.7919, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(569.2485, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(957.9145, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1251.6206, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1298.0742, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1007.4423, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(905.4270, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(930.7674, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(640.0196, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(673.9435, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(558.0591, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(195.2205, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(514.6820, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(448.1400, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(925.2326, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1062.5396, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1276.3687, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(813.0511, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(904.1116, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(837.8412, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(869.0240, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1130.2021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(482.7867, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(757.7595, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(673.4767, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(394.8169, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(552.2118, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(647.4579, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(978.3807, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1252.0592, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1610.4309, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.1756198347107438, 'recall': 0.21170610211706103, 'f1': 0.1919819311123659, 'number': 803}, 'P': {'precision': 0.3143491124260355, 'recall': 0.5145278450363197, 'f1': 0.3902662993572085, 'number': 826}, 'overall_precision': 0.25646551724137934, 'overall_recall': 0.3652547575199509, 'overall_f1': 0.3013421119270702, 'overall_accuracy': 0.6261156383391541}
			------------EPOCH 8---------------
Loss:  tensor(977.7737, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(772.1698, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1081.1813, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(551.1993, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(841.0667, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1011.3978, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1277.8329, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(913.6221, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(904.3107, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1035.1775, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(754.1007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(715.0754, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(722.1450, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(326.1706, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(525.2003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(619.4629, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1140.8904, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1320.5793, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1479.3752, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1083.3845, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(926.6489, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(894.6343, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(945.1075, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1285.4036, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(531.9886, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(778.3885, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(786.7460, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(325.7972, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(355.7983, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(473.9351, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(710.1687, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1052.9060, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1168.6752, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.15865384615384615, 'recall': 0.0821917808219178, 'f1': 0.10828547990155865, 'number': 803}, 'P': {'precision': 0.4035936420179682, 'recall': 0.7070217917675545, 'f1': 0.5138583369995601, 'number': 826}, 'overall_precision': 0.3488996242619431, 'overall_recall': 0.39901780233271944, 'overall_f1': 0.37227949599083615, 'overall_accuracy': 0.6344586728754366}
			------------EPOCH 9---------------
Loss:  tensor(952.8763, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(856.6769, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(885.5266, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(531.5735, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(842.7280, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1144.3967, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1794.4795, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(953.0097, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(894.7559, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(930.3818, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(428.7338, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(750.4039, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(530.5070, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(175.4039, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(534.1888, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(268.4165, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(593.8484, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(692.7315, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(841.2734, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(526.2649, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(582.0519, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(548.1124, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(492.2220, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(696.1129, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(340.1389, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(441.0892, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(544.8250, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(260.2146, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(236.4035, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(309.3217, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(568.0994, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(881.3330, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1235.8536, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.2570605187319885, 'recall': 0.5554171855541719, 'f1': 0.3514578408195429, 'number': 803}, 'P': {'precision': 0.4878640776699029, 'recall': 0.24334140435835352, 'f1': 0.3247172859450727, 'number': 826}, 'overall_precision': 0.30135072193758733, 'overall_recall': 0.39717618170656843, 'overall_f1': 0.3426906779661017, 'overall_accuracy': 0.6253395421032208}
			------------EPOCH 10---------------
Loss:  tensor(641.1602, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(434.7354, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(698.3422, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(428.8322, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(630.0156, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(669.4948, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(908.9144, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(665.1799, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(545.7078, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(502.3132, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(335.4400, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(383.3602, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(290.8898, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(109.5767, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(290.5556, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(160.2083, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(396.6393, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(399.7001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(527.4470, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(326.9118, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(523.6187, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(510.1151, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(356.6278, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(559.2532, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(293.2146, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(376.7663, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(429.3082, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(220.6585, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(283.1822, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(219.6389, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(458.1315, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(458.3042, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(607.0995, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.2930817610062893, 'recall': 0.2901618929016189, 'f1': 0.2916145181476846, 'number': 803}, 'P': {'precision': 0.43634840871021774, 'recall': 0.6307506053268765, 'f1': 0.5158415841584157, 'number': 826}, 'overall_precision': 0.3790849673202614, 'overall_recall': 0.46286065070595456, 'overall_f1': 0.4168048645660586, 'overall_accuracy': 0.6887126503686457}
			------------EPOCH 11---------------
Loss:  tensor(433.7248, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(273.7072, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(329.4341, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(261.1303, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(325.6290, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(426.1155, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(682.2743, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(446.6740, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(375.8184, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(362.6951, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(200.4501, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(296.7757, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(219.4863, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(99.7568, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(251.6498, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(127.5143, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(320.3796, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(362.4677, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(488.7593, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(259.3036, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(459.3472, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(394.4533, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(279.0344, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(410.8398, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(194.9799, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(209.4321, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(335.3967, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(137.2656, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(133.4632, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(135.5480, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(304.9488, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(368.1536, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(434.7660, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.27040816326530615, 'recall': 0.3300124533001245, 'f1': 0.2972518227706113, 'number': 803}, 'P': {'precision': 0.4722222222222222, 'recall': 0.6174334140435835, 'f1': 0.5351521511017838, 'number': 826}, 'overall_precision': 0.3762135922330097, 'overall_recall': 0.47575199508901167, 'overall_f1': 0.4201680672268908, 'overall_accuracy': 0.6984623593325573}
			------------EPOCH 12---------------
Loss:  tensor(311.8797, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(173.7493, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(277.8908, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(199.8376, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(244.6717, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(271.3450, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(489.5745, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(317.1909, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(319.4576, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(290.1504, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(135.8202, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(238.6808, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(191.4557, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.4836, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(211.7328, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(89.0374, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(217.6354, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(227.1620, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(350.2178, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(215.8678, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(377.6998, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(238.7567, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(177.1356, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(240.2747, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(138.1050, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(143.6512, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(207.9407, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(109.7461, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(81.1555, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(84.9507, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(195.8332, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(218.0517, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(330.3763, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.32571428571428573, 'recall': 0.49688667496886674, 'f1': 0.39349112426035504, 'number': 803}, 'P': {'precision': 0.49206349206349204, 'recall': 0.48789346246973364, 'f1': 0.4899696048632219, 'number': 826}, 'overall_precision': 0.3923679060665362, 'overall_recall': 0.4923265807243708, 'overall_f1': 0.43670024503130955, 'overall_accuracy': 0.6786476523088864}
			------------EPOCH 13---------------
Loss:  tensor(283.8808, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(115.2060, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(178.8899, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(225.8674, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(170.7065, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(170.3506, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(361.5783, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(255.2829, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(247.4217, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(175.7939, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(85.3250, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(154.3476, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(117.0759, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.7167, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(115.9325, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.9233, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(143.5430, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(121.9106, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(178.2986, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(110.9470, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(283.9873, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(171.0573, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(133.1900, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(171.2591, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(98.1131, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(106.6291, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(151.1539, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(100.0309, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(81.5342, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(83.8191, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(161.6559, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(130.7071, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(172.2936, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3181818181818182, 'recall': 0.33997509339975096, 'f1': 0.3287176399759182, 'number': 803}, 'P': {'precision': 0.4418394950405771, 'recall': 0.5932203389830508, 'f1': 0.5064599483204135, 'number': 826}, 'overall_precision': 0.3879003558718861, 'overall_recall': 0.46838551258440764, 'overall_f1': 0.42436040044493883, 'overall_accuracy': 0.7011301901435778}
			------------EPOCH 14---------------
Loss:  tensor(160.8048, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(89.8013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(93.0818, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(103.5308, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(109.7634, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(89.4079, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(188.6008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(161.4876, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(174.7374, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(105.6854, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(59.6585, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(122.4932, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(112.6497, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.7405, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(98.6808, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.7625, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(119.0790, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(117.1376, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(134.7944, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(75.6964, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(190.4551, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(117.6547, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(86.3987, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(165.5663, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(75.9403, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(67.1262, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(109.2134, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(67.0955, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.8797, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.4294, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(92.4970, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(81.9903, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(93.3949, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3398843930635838, 'recall': 0.36612702366127026, 'f1': 0.35251798561151076, 'number': 803}, 'P': {'precision': 0.4730077120822622, 'recall': 0.6682808716707022, 'f1': 0.5539387857501253, 'number': 826}, 'overall_precision': 0.41633858267716534, 'overall_recall': 0.5193370165745856, 'overall_f1': 0.4621688063370664, 'overall_accuracy': 0.6980985642219635}
			------------EPOCH 15---------------
Loss:  tensor(94.7778, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(82.8790, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(68.5887, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(85.3144, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(81.9852, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(71.0585, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(168.7747, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(144.2368, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(172.5781, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(88.1030, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.9743, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(107.2871, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(83.3474, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.3440, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(61.5720, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.7316, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(80.5708, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.9871, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(90.6457, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.1740, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(142.1645, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(86.5848, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(71.1395, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(95.1722, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.2562, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.3369, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(82.6717, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(59.7648, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.8939, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.3415, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(81.5463, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(83.7460, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(110.3636, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.2786026200873362, 'recall': 0.3972602739726027, 'f1': 0.32751540041067756, 'number': 803}, 'P': {'precision': 0.4630593132154006, 'recall': 0.5387409200968523, 'f1': 0.498041410184667, 'number': 826}, 'overall_precision': 0.3627730294396961, 'overall_recall': 0.468999386126458, 'overall_f1': 0.409103078982597, 'overall_accuracy': 0.686651144741948}
			------------EPOCH 16---------------
Loss:  tensor(96.4035, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(73.0363, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(61.2329, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(74.4568, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(75.5326, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.3255, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(139.8453, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(99.6174, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(117.3019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(69.0919, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.0178, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(107.6854, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(90.3207, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.8606, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(68.5616, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.7478, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(83.2281, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.4554, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(83.3942, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.2861, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(109.6774, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(87.7383, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(61.3576, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(94.4458, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.4221, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.8721, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(66.5982, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.0168, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.4551, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.2120, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(70.2799, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(66.4914, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(82.9569, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3675115207373272, 'recall': 0.3972602739726027, 'f1': 0.3818073010173548, 'number': 803}, 'P': {'precision': 0.4556848701880036, 'recall': 0.6162227602905569, 'f1': 0.5239320638188368, 'number': 826}, 'overall_precision': 0.4171284634760705, 'overall_recall': 0.5082872928176796, 'overall_f1': 0.4582180409518539, 'overall_accuracy': 0.7059565386107878}
			------------EPOCH 17---------------
Loss:  tensor(81.3169, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(69.0686, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.8764, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.0691, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.2791, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.0313, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(97.7237, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(66.3555, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(73.9098, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.9053, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.7702, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(75.6924, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.2020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.8521, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.4472, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.5836, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.7464, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.7406, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.0792, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.5200, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(64.5129, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.2107, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.9710, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(79.4530, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.1138, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.0900, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.2041, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.3419, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.7044, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.2304, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.6745, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.9335, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.6826, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.37446808510638296, 'recall': 0.4383561643835616, 'f1': 0.40390131956397013, 'number': 803}, 'P': {'precision': 0.4821944177093359, 'recall': 0.6065375302663438, 'f1': 0.5372654155495977, 'number': 826}, 'overall_precision': 0.43102577059120767, 'overall_recall': 0.523634131368938, 'overall_f1': 0.4728381374722838, 'overall_accuracy': 0.6981713232440823}
			------------EPOCH 18---------------
Loss:  tensor(60.7845, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.4846, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.3114, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.0373, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.9789, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.9342, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.9690, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.2018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.5405, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.1986, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.4514, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.4104, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.4663, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.8412, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.6336, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.5785, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.1801, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.4183, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.7107, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.7583, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.9237, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.1746, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.3760, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.4638, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.6822, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.5712, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.7010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.3207, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.7779, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.9721, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.9271, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.7938, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.2058, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3806941431670282, 'recall': 0.43711083437110837, 'f1': 0.40695652173913044, 'number': 803}, 'P': {'precision': 0.4851104707012488, 'recall': 0.6113801452784504, 'f1': 0.5409748259239422, 'number': 826}, 'overall_precision': 0.4360672440142639, 'overall_recall': 0.525475751995089, 'overall_f1': 0.4766146993318486, 'overall_accuracy': 0.7011544431509508}
			------------EPOCH 19---------------
Loss:  tensor(45.0012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.5591, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.5044, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.9793, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.5666, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.6391, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.2663, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.0447, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.4883, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.3581, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.1538, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.5990, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.6767, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.2605, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.3746, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.5731, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.7813, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.4132, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.8291, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.8117, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.2613, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.8226, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.2516, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.5640, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.4693, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.0746, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.6403, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.1351, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.1011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.3285, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.3486, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.9296, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.4639, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.388641425389755, 'recall': 0.43462017434620176, 'f1': 0.4103468547912993, 'number': 803}, 'P': {'precision': 0.48861480075901326, 'recall': 0.6234866828087167, 'f1': 0.5478723404255319, 'number': 826}, 'overall_precision': 0.4426229508196721, 'overall_recall': 0.5303867403314917, 'overall_f1': 0.4825467746439542, 'overall_accuracy': 0.702197322467986}
			------------EPOCH 20---------------
Loss:  tensor(30.2792, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.6273, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.0027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.3962, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.9751, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.2561, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.7701, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.9908, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.0662, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.2935, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.1822, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.5986, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.0157, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.6512, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.4501, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.6547, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.4638, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.0583, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.7898, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.7354, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.6864, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.9251, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.1601, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.8329, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.2145, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.3773, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.7505, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.0111, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.4014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.0467, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.2335, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.2727, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.4243, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.39976958525345624, 'recall': 0.43212951432129515, 'f1': 0.41532016756433277, 'number': 803}, 'P': {'precision': 0.49097815764482433, 'recall': 0.62590799031477, 'f1': 0.5502927088877062, 'number': 826}, 'overall_precision': 0.4497657470067673, 'overall_recall': 0.5303867403314917, 'overall_f1': 0.4867605633802817, 'overall_accuracy': 0.7033857198292588}
			------------EPOCH 21---------------
Loss:  tensor(22.1378, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.0815, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.5589, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.2347, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.7740, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.7719, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.0994, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.5467, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.5034, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.4157, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.1524, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.2188, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.9203, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.8245, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.3246, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.5370, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.9239, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.2749, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.5752, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.8363, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.4592, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.2717, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.4363, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.0842, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.8933, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.9283, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.1829, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.0839, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.3813, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.0349, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.0737, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.8409, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.9637, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3795698924731183, 'recall': 0.4396014943960149, 'f1': 0.40738603577611077, 'number': 803}, 'P': {'precision': 0.4595348837209302, 'recall': 0.5980629539951574, 'f1': 0.5197264597580221, 'number': 826}, 'overall_precision': 0.4224438902743142, 'overall_recall': 0.519950890116636, 'overall_f1': 0.4661529994496422, 'overall_accuracy': 0.6985836243694218}
			------------EPOCH 22---------------
Loss:  tensor(25.5969, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.7087, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.6998, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.9992, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.0949, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.8826, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.1252, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.2648, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.6322, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.1228, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.2932, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.8330, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.2850, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.5984, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.8196, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.2018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.7449, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.9777, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.8986, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.6581, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.4698, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.1826, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.1303, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.3830, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.9227, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.6641, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.2757, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.5613, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.8923, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.9424, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.3005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.3458, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.2875, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3866213151927438, 'recall': 0.4246575342465753, 'f1': 0.4047477744807122, 'number': 803}, 'P': {'precision': 0.4606116774791474, 'recall': 0.6016949152542372, 'f1': 0.5217847769028872, 'number': 826}, 'overall_precision': 0.4273329933707292, 'overall_recall': 0.514426028238183, 'overall_f1': 0.4668523676880223, 'overall_accuracy': 0.7022943344974777}
			------------EPOCH 23---------------
Loss:  tensor(17.7458, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.8449, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.2831, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.9312, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.4320, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.7345, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.0104, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.5275, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.0619, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.7684, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.5801, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.6990, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.8117, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.9899, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.3354, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.1043, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.7465, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.5143, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.3639, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.3612, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.7443, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.6062, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.3137, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.8200, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.1587, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.4181, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.1131, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.9951, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.7951, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.9952, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.7683, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.6910, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.7145, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.40765765765765766, 'recall': 0.45080946450809467, 'f1': 0.4281490242460083, 'number': 803}, 'P': {'precision': 0.45445736434108525, 'recall': 0.5677966101694916, 'f1': 0.5048439181916039, 'number': 826}, 'overall_precision': 0.4328125, 'overall_recall': 0.5101289134438306, 'overall_f1': 0.46830092983939137, 'overall_accuracy': 0.7045741171905316}
			------------EPOCH 24---------------
Loss:  tensor(12.0809, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.5769, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.3833, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.8506, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.6594, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.8409, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.4382, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.2980, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.7538, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.6073, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.2115, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.6339, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.6558, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.4206, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.9994, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.5732, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.5403, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.6708, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.7222, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.2873, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.9182, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.4626, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.0900, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.1608, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.4292, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.9363, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.2965, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.6258, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.1105, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.6059, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.5378, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.2617, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.9988, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3988372093023256, 'recall': 0.42714819427148193, 'f1': 0.41250751653638007, 'number': 803}, 'P': {'precision': 0.47368421052631576, 'recall': 0.6537530266343826, 'f1': 0.5493387589013224, 'number': 826}, 'overall_precision': 0.4415, 'overall_recall': 0.5420503376304481, 'overall_f1': 0.4866354367594379, 'overall_accuracy': 0.7029734187039193}
			------------EPOCH 25---------------
Loss:  tensor(9.9682, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.2288, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.0695, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.3786, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.9301, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.7866, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.5140, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.4578, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.1252, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.3491, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.7428, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.0160, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.8206, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.7932, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.2536, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.2033, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.6074, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.4299, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.1534, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.7206, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.7736, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.7704, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.0702, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.2426, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.6716, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.0330, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.1318, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.4205, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.1561, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.3917, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.8952, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.6218, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.7141, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4102272727272727, 'recall': 0.44956413449564137, 'f1': 0.4289958407605467, 'number': 803}, 'P': {'precision': 0.47297297297297297, 'recall': 0.5932203389830508, 'f1': 0.5263157894736841, 'number': 826}, 'overall_precision': 0.4441544885177453, 'overall_recall': 0.5224063842848373, 'overall_f1': 0.48011283497884344, 'overall_accuracy': 0.7074117190531626}
			------------EPOCH 26---------------
Loss:  tensor(6.6389, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.6305, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.3076, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.1181, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.5454, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.3207, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.2356, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.0995, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.3021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.8203, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.2013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.6683, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.3319, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5773, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.9755, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5742, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.3566, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.7897, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.1145, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.8050, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.0385, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.6458, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.4206, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.4017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.1509, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.1594, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.7638, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.8056, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.0616, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.7268, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.3782, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.5526, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.7580, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4205298013245033, 'recall': 0.47447073474470736, 'f1': 0.44587478057343477, 'number': 803}, 'P': {'precision': 0.4838403041825095, 'recall': 0.6162227602905569, 'f1': 0.5420660276890308, 'number': 826}, 'overall_precision': 0.45454545454545453, 'overall_recall': 0.5463474524248005, 'overall_f1': 0.49623640925564544, 'overall_accuracy': 0.7024641055490881}
			------------EPOCH 27---------------
Loss:  tensor(5.8782, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.7183, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.6409, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.9848, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.1182, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.7699, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.6426, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.9805, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.7588, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.2473, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.3109, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.7763, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.4102, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3675, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.9423, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2205, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.5653, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.8766, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.2369, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.9979, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.4394, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.2406, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.2434, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.5723, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5311, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.6080, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.3791, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.9646, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.9557, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.0434, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.3447, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.5263, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.8730, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4183082271147161, 'recall': 0.44956413449564137, 'f1': 0.4333733493397359, 'number': 803}, 'P': {'precision': 0.4780952380952381, 'recall': 0.6077481840193705, 'f1': 0.5351812366737739, 'number': 826}, 'overall_precision': 0.4511238891792995, 'overall_recall': 0.5297728667894414, 'overall_f1': 0.48729531338226983, 'overall_accuracy': 0.7074844780752814}
			------------EPOCH 28---------------
Loss:  tensor(4.5431, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.5031, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.1844, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.6529, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.0567, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.7069, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.5760, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.6811, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.2775, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.3680, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.7292, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.8932, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.8505, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1142, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.4245, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.8886, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.0244, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.2418, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.7568, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.5335, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.5318, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.6479, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.6632, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.0373, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2255, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.2862, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.8250, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.9189, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.7461, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.6728, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.8321, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.8020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.3254, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4215349369988545, 'recall': 0.45828144458281445, 'f1': 0.43914081145584727, 'number': 803}, 'P': {'precision': 0.4748603351955307, 'recall': 0.6174334140435835, 'f1': 0.5368421052631579, 'number': 826}, 'overall_precision': 0.4509501797637391, 'overall_recall': 0.5389809699201964, 'overall_f1': 0.49105145413870244, 'overall_accuracy': 0.7052532013969732}
			------------EPOCH 29---------------
Loss:  tensor(3.9635, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.4140, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.3221, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.6999, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.2799, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.2116, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.1512, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.6197, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.9161, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.2046, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.3114, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.0366, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.2613, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0664, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.0147, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.7317, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.5795, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.9896, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.2689, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.3108, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.6958, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.3342, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.8172, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.6095, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.0325, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.0026, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.4570, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.7479, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6089, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.4323, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.4232, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.4090, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.8889, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4235159817351598, 'recall': 0.46201743462017436, 'f1': 0.44192972007147113, 'number': 803}, 'P': {'precision': 0.4738344433872502, 'recall': 0.6029055690072639, 'f1': 0.5306339904102291, 'number': 826}, 'overall_precision': 0.4509600415153088, 'overall_recall': 0.5334561080417434, 'overall_f1': 0.48875140607424067, 'overall_accuracy': 0.7021488164532402}
			------------EPOCH 30---------------
Loss:  tensor(3.4348, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.6899, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.9001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.8324, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.6553, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.5701, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.6744, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.6017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.2683, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.4758, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.0497, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.2559, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.8241, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9268, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.6562, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5167, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.1141, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.5095, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.9152, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.8721, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.3458, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.8848, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.4136, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.1968, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.8305, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.7060, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.9958, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.7478, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4152, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.1963, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.0559, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.1558, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.5548, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.42479908151549944, 'recall': 0.46077210460772106, 'f1': 0.44205495818399043, 'number': 803}, 'P': {'precision': 0.47423664122137404, 'recall': 0.6016949152542372, 'f1': 0.5304162219850587, 'number': 826}, 'overall_precision': 0.4517978113600834, 'overall_recall': 0.5322283609576427, 'overall_f1': 0.4887260428410372, 'overall_accuracy': 0.7039192859914629}


		-------------RUN 2-----------
			------------EPOCH 1---------------
Loss:  tensor(1531.2037, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1460.7065, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2539.9697, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2820.0127, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2967.0864, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2371.8442, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2053.8828, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2450.0513, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1627.8138, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2433.3179, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1966.4851, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1301.8418, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2547.4946, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1914.1588, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2921.2695, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2247.0693, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2815.3696, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1203.9438, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1922.2631, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2313.0735, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1994.1210, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2069.7007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1335.5697, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1302.7114, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(797.5427, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1365.9200, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1573.6886, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2221.1875, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2089.8381, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1507.7205, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1852.7534, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1843.4366, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2494.6260, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2079.8291, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1222.8416, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'number': 820}, 'P': {'precision': 0.14527687296416938, 'recall': 0.25312145289443816, 'f1': 0.18460264900662252, 'number': 881}, 'overall_precision': 0.14433656957928803, 'overall_recall': 0.13109935332157555, 'overall_f1': 0.13739987677141097, 'overall_accuracy': 0.4229833217977628}
			------------EPOCH 2---------------
Loss:  tensor(1024.9707, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(986.6505, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1682.6373, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1955.0026, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2140.1477, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1728.8300, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1470.8291, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1738.5261, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1222.1016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1860.9204, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1468.2573, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(967.2592, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2078.1841, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1484.0762, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2327.3154, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1679.8262, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2324.2136, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(944.8748, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1563.9991, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1821.5830, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1616.8330, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1743.8706, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1103.6243, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1094.2673, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(640.5648, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1172.7477, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1273.8569, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1990.0635, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1693.7444, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1254.5818, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1574.9020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1582.1377, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2046.4347, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1836.8469, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(969.0984, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.01461038961038961, 'recall': 0.01097560975609756, 'f1': 0.012534818941504178, 'number': 820}, 'P': {'precision': 0.26064178127046495, 'recall': 0.4517593643586833, 'f1': 0.33056478405315615, 'number': 881}, 'overall_precision': 0.18992067195520299, 'overall_recall': 0.23927101704879483, 'overall_f1': 0.21175858480749218, 'overall_accuracy': 0.5067762173748018}
			------------EPOCH 3---------------
Loss:  tensor(829.9358, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(732.4587, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1300.0828, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1594.2983, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1734.9457, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1331.6051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1123.0510, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1424.2007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1007.9278, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1601.7316, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1192.4658, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(735.9894, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1876.8638, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1273.7913, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1807.6339, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1321.3978, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1891.6868, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(746.6266, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1102.0365, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1411.2888, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1300.8862, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1574.7899, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(914.5537, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(917.8406, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(546.6533, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(978.9443, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(936.5359, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1695.7234, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1415.4897, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1052.4447, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1356.3210, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1382.4055, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1492.2235, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1761.0830, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(758.0089, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.06784968684759916, 'recall': 0.07926829268292683, 'f1': 0.07311586051743532, 'number': 820}, 'P': {'precision': 0.2903010033444816, 'recall': 0.492622020431328, 'f1': 0.3653198653198653, 'number': 881}, 'overall_precision': 0.2034243783122707, 'overall_recall': 0.2933568489124045, 'overall_f1': 0.24025036109773712, 'overall_accuracy': 0.558105784902543}
			------------EPOCH 4---------------
Loss:  tensor(735.2751, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(586.5745, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1018.4443, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1324.5208, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1422.3516, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(975.6646, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(851.5676, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1230.8384, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(808.6299, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1290.4194, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(993.5110, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(565.6965, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1467.4438, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(988.6271, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1375.0774, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1019.2256, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1553.6616, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(608.2524, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(775.2078, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1176.1918, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(935.6211, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1322.6531, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(771.2172, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(761.2815, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(406.1408, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(712.8333, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(649.6241, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1297.0538, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(987.4389, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(816.8533, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(959.7095, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1161.3245, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1148.9338, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1235.8755, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(622.6407, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.13076923076923078, 'recall': 0.12439024390243902, 'f1': 0.1275, 'number': 820}, 'P': {'precision': 0.2922794117647059, 'recall': 0.5414301929625426, 'f1': 0.3796259450855551, 'number': 881}, 'overall_precision': 0.24004975124378108, 'overall_recall': 0.3403880070546737, 'overall_f1': 0.2815463165572574, 'overall_accuracy': 0.5707204894058809}
			------------EPOCH 5---------------
Loss:  tensor(611.5981, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(422.0656, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(788.3360, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1069.3220, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1036.6323, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(590.0766, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(524.3756, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(972.1765, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(684.9850, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1040.5466, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1019.5907, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(608.0314, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1078.2841, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(736.9698, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1050.7764, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(773.4297, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1117.4894, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(482.1465, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(513.8756, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(963.4821, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(667.4429, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1299.9958, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(593.3350, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(498.2050, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(261.0097, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(536.6050, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(451.0851, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(900.9561, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(742.4932, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(637.7478, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(737.1069, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(898.8071, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(834.6033, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(856.1781, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(516.6324, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.14182692307692307, 'recall': 0.14390243902439023, 'f1': 0.14285714285714288, 'number': 820}, 'P': {'precision': 0.331421287444232, 'recall': 0.5902383654937571, 'f1': 0.42448979591836733, 'number': 881}, 'overall_precision': 0.26572261557684296, 'overall_recall': 0.3750734861845973, 'overall_f1': 0.31106777181862505, 'overall_accuracy': 0.5994775502913662}
			------------EPOCH 6---------------
Loss:  tensor(464.3616, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(282.6347, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(587.8719, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(782.5215, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(781.2802, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(358.7310, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(330.0497, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(717.4897, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(569.9766, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(829.1616, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(932.0005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(364.3180, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(783.1411, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(544.6775, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(788.0590, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(603.8548, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1263.1780, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(452.4347, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(602.0424, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1123.3535, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(444.5948, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(982.0961, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(427.7546, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(320.7182, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(165.4701, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(321.2707, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(303.7791, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(584.0356, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(717.7338, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(632.8903, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(770.7504, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(771.8696, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(664.3421, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1007.9587, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(492.9893, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.21343283582089553, 'recall': 0.174390243902439, 'f1': 0.1919463087248322, 'number': 820}, 'P': {'precision': 0.27599557522123896, 'recall': 0.5664018161180476, 'f1': 0.3711416883599851, 'number': 881}, 'overall_precision': 0.25907990314769974, 'overall_recall': 0.37742504409171074, 'overall_f1': 0.30725053840631733, 'overall_accuracy': 0.5532831722074617}
			------------EPOCH 7---------------
Loss:  tensor(476.5832, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(280.0872, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(590.7673, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(646.3602, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(652.7808, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(288.4892, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(253.8496, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(687.0425, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(457.2217, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(507.5023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(521.4462, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(315.3703, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(545.9237, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(406.5372, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(653.1677, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(465.5139, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(707.8828, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(247.5777, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(268.6417, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(664.1464, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(637.4122, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1014.9573, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(575.7012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(407.9727, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(151.6455, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(387.3704, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(529.7212, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(716.0892, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(652.9901, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(507.9908, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(592.8539, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(697.1384, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(449.7667, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(398.4047, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(386.7079, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.23374613003095976, 'recall': 0.18414634146341463, 'f1': 0.20600272851296045, 'number': 820}, 'P': {'precision': 0.29739130434782607, 'recall': 0.5822928490351873, 'f1': 0.39370683039140447, 'number': 881}, 'overall_precision': 0.28005061155630534, 'overall_recall': 0.3903586125808348, 'overall_f1': 0.3261296660117878, 'overall_accuracy': 0.5767264283641073}
			------------EPOCH 8---------------
Loss:  tensor(291.5226, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(194.2431, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(375.7893, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(454.7457, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(663.2738, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(349.2430, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(311.9051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(669.9945, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(713.9223, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(933.2949, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1138.6138, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(723.2734, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1081.0416, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(601.7257, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(979.7245, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(641.7562, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(906.7416, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(251.2890, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(417.9904, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(625.7573, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(354.9324, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(998.1364, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(428.5012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(314.9004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(153.3907, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(333.7044, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(298.0534, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(446.6568, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(503.9333, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(398.2838, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(565.8839, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(830.9896, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1006.2603, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(667.1557, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(688.8452, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.12767780634104542, 'recall': 0.18170731707317073, 'f1': 0.14997483643683945, 'number': 820}, 'P': {'precision': 0.41569767441860467, 'recall': 0.4869466515323496, 'f1': 0.44851019341348675, 'number': 881}, 'overall_precision': 0.2628467485220555, 'overall_recall': 0.33980011757789536, 'overall_f1': 0.29641025641025637, 'overall_accuracy': 0.6301770524012592}
			------------EPOCH 9---------------
Loss:  tensor(589.8206, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(289.4845, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(610.2615, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(791.6460, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(670.7384, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(456.4011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(347.5673, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(686.7360, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(444.0703, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(470.3024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(612.6488, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(393.2821, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(629.6083, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(382.2016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(585.6805, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(454.5549, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(896.8172, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(373.3085, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(401.7667, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(916.4752, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(437.0041, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(840.8728, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(321.4193, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(208.8255, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(123.3092, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(177.1965, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(202.0201, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(388.3439, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(363.6761, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(313.1273, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(420.2947, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(320.6237, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(288.8149, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(646.8304, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(280.0062, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.26525017135023987, 'recall': 0.4719512195121951, 'f1': 0.33962264150943394, 'number': 820}, 'P': {'precision': 0.39399624765478425, 'recall': 0.4767309875141884, 'f1': 0.43143297380585516, 'number': 881}, 'overall_precision': 0.3196039603960396, 'overall_recall': 0.47442680776014107, 'overall_f1': 0.3819214387127307, 'overall_accuracy': 0.6136551385384804}
			------------EPOCH 10---------------
Loss:  tensor(343.5189, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(214.0746, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(489.2764, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(570.4705, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(683.1855, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(463.9700, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(264.6362, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(672.0563, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(347.4710, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(322.7153, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(404.0251, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(254.0441, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(271.4655, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(192.9240, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(389.2179, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(265.4941, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(461.4230, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(174.9413, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(155.8701, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(409.2900, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(192.9642, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(592.1833, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(234.0300, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(158.2827, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(133.1228, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(203.5596, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(179.0208, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(347.9226, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(408.6995, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(424.3175, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(467.1995, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(461.9376, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(260.7310, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(404.5083, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(442.2719, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.2689210950080515, 'recall': 0.20365853658536584, 'f1': 0.23178348369188065, 'number': 820}, 'P': {'precision': 0.4174757281553398, 'recall': 0.6833144154370034, 'f1': 0.5182953077916488, 'number': 881}, 'overall_precision': 0.3727581192438197, 'overall_recall': 0.4520870076425632, 'overall_f1': 0.4086078639744952, 'overall_accuracy': 0.6450914286990109}
			------------EPOCH 11---------------
Loss:  tensor(149.2626, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(109.0291, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(219.9276, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(321.9608, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(307.4871, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(95.3774, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(88.3538, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(266.8391, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(195.6531, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(226.7353, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(208.8166, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(175.0719, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(307.2658, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(197.4021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(390.4587, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(267.0941, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(364.7781, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(149.0866, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(147.6696, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(369.8552, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(97.7560, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(317.9745, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(150.2832, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(118.7366, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(66.3361, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(104.4207, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(108.0896, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(167.2432, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(179.3568, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(137.3862, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(235.7604, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(150.7195, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(119.4474, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(145.0784, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(146.8033, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.27926960257787325, 'recall': 0.3170731707317073, 'f1': 0.296973158195317, 'number': 820}, 'P': {'precision': 0.42487046632124353, 'recall': 0.6515323496027242, 'f1': 0.5143369175627239, 'number': 881}, 'overall_precision': 0.3654688869412796, 'overall_recall': 0.49029982363315694, 'overall_f1': 0.4187798142103942, 'overall_accuracy': 0.6580410368617294}
			------------EPOCH 12---------------
Loss:  tensor(91.8029, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(71.2497, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(149.1821, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(221.5080, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(230.6344, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.8098, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.2092, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(178.6599, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(120.9913, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(147.4370, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(158.0870, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(125.2521, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(143.6039, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(88.4073, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(221.1536, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(139.9960, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(225.3199, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(79.9751, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(70.3220, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(189.3994, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.0395, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(196.6208, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(114.2605, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(89.6962, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.8245, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(71.6511, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(81.1468, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(144.2558, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(118.7093, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(93.0156, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(151.8809, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(110.4192, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(83.1743, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(105.5356, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(114.0784, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3296398891966759, 'recall': 0.29024390243902437, 'f1': 0.3086900129701686, 'number': 820}, 'P': {'precision': 0.4267241379310345, 'recall': 0.674233825198638, 'f1': 0.5226572811262649, 'number': 881}, 'overall_precision': 0.3935666982024598, 'overall_recall': 0.48912404467960025, 'overall_f1': 0.436173001310616, 'overall_accuracy': 0.6520127709928777}
			------------EPOCH 13---------------
Loss:  tensor(52.8070, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.8912, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(106.7974, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(194.1249, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(186.0692, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.4044, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.2641, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(116.8525, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(74.5412, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(109.8676, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(95.4438, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(79.8226, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(90.0095, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(67.4553, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(127.5901, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(107.1133, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(154.3110, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.9178, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(68.4917, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(130.5348, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.4003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(129.5202, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(74.0745, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(79.0751, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.1948, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.7575, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(69.2504, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(86.5092, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(92.6531, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(71.4562, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(126.9522, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(93.7820, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(64.5429, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(65.9365, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(84.5678, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.32823529411764707, 'recall': 0.3402439024390244, 'f1': 0.33413173652694617, 'number': 820}, 'P': {'precision': 0.4642857142857143, 'recall': 0.6640181611804767, 'f1': 0.5464736104624008, 'number': 881}, 'overall_precision': 0.409478672985782, 'overall_recall': 0.5079365079365079, 'overall_f1': 0.45342429808449225, 'overall_accuracy': 0.6646274754962156}
			------------EPOCH 14---------------
Loss:  tensor(37.6469, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.0808, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(85.6447, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(157.1190, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(149.7388, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.0570, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.3587, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(81.9625, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.4990, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(88.2147, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(73.0870, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.4869, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(70.4853, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.2940, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(78.1046, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(86.8599, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(116.3719, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.7851, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.5524, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(104.0794, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.4352, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(97.9366, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.5486, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(70.6764, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.8755, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.4677, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.8907, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.9891, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(59.4351, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.4192, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(88.4340, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(67.7944, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.9546, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.6782, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.9770, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3633986928104575, 'recall': 0.33902439024390246, 'f1': 0.35078864353312306, 'number': 820}, 'P': {'precision': 0.4496996996996997, 'recall': 0.6799091940976163, 'f1': 0.5413465883416178, 'number': 881}, 'overall_precision': 0.41821649976156416, 'overall_recall': 0.5155790711346266, 'overall_f1': 0.46182201158504477, 'overall_accuracy': 0.6653419366362276}
			------------EPOCH 15---------------
Loss:  tensor(31.3371, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.6236, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.1700, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(125.9879, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(125.9419, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.6661, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.7184, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(65.9569, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.3002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(75.9992, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(62.7225, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.5517, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.0455, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.0662, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(62.5465, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(75.1351, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(100.9017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.8838, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.1732, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(81.3344, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.1956, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(81.4299, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.2656, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(64.7878, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.5194, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.9258, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.9044, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.3664, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.4771, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.6456, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(70.3044, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.9561, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.0846, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.4305, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.8923, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3806970509383378, 'recall': 0.3463414634146341, 'f1': 0.3627075351213282, 'number': 820}, 'P': {'precision': 0.456968773800457, 'recall': 0.681044267877412, 'f1': 0.5469462169553326, 'number': 881}, 'overall_precision': 0.4293346284604177, 'overall_recall': 0.5196942974720753, 'overall_f1': 0.4702127659574468, 'overall_accuracy': 0.6654312442787291}
			------------EPOCH 16---------------
Loss:  tensor(27.6361, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.9497, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.5263, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(99.9367, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(105.8722, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.9262, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.6057, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.3300, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.0243, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(64.3411, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.9280, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.2045, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.7814, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.3637, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.7570, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(64.0339, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(86.8767, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.4831, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.5151, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(66.3152, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.0061, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(71.4329, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.6369, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(59.0230, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.8135, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.3596, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.1944, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.2416, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.6070, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.6606, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.5363, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.1698, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.2604, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.5873, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.6076, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3820078226857888, 'recall': 0.3573170731707317, 'f1': 0.3692501575299307, 'number': 820}, 'P': {'precision': 0.46966115051221435, 'recall': 0.6765039727582293, 'f1': 0.5544186046511629, 'number': 881}, 'overall_precision': 0.43664047151277013, 'overall_recall': 0.522633744855967, 'overall_f1': 0.4757827134064758, 'overall_accuracy': 0.6679318582687713}
			------------EPOCH 17---------------
Loss:  tensor(24.6738, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.1511, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.2852, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(77.0723, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(89.6201, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.3509, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.6552, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.4632, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.7422, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.6855, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.6348, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.6714, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.4112, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.8344, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.3247, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.8542, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(79.2957, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.5031, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.5835, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.5103, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.3517, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(61.6959, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.3550, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.0372, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.3453, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.3729, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.2893, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.3925, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.7345, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.4731, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.1325, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.9163, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.9579, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.5203, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.7960, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3821571238348868, 'recall': 0.35, 'f1': 0.36537237428389563, 'number': 820}, 'P': {'precision': 0.46420323325635104, 'recall': 0.684449489216799, 'f1': 0.553211009174312, 'number': 881}, 'overall_precision': 0.43414634146341463, 'overall_recall': 0.5232216343327455, 'overall_f1': 0.4745401226339642, 'overall_accuracy': 0.6662796668824935}
			------------EPOCH 18---------------
Loss:  tensor(22.3205, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.6765, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.1286, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(62.3990, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(79.1620, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.7664, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.3293, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.6687, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.8808, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.2182, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.7641, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.8934, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.5435, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.5053, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.2348, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.6761, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(71.7332, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.7579, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.2615, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.1558, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.9133, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.5587, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.5844, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.3129, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.7900, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.6747, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.5431, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.4449, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.0984, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.6470, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.7581, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.2980, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.4514, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.7491, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.2970, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3900134952766532, 'recall': 0.3524390243902439, 'f1': 0.370275464445868, 'number': 820}, 'P': {'precision': 0.4625382262996942, 'recall': 0.6867196367763905, 'f1': 0.5527638190954773, 'number': 881}, 'overall_precision': 0.4363103953147877, 'overall_recall': 0.5255731922398589, 'overall_f1': 0.4768, 'overall_accuracy': 0.6659670901337382}
			------------EPOCH 19---------------
Loss:  tensor(20.3600, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.3415, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.2074, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.7073, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(70.9059, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.9039, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.4513, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.7410, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.3661, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.8990, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.3523, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.4603, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.8247, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.1651, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.5748, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.2847, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(68.3409, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.0843, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.9913, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.3791, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.2474, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.8693, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.5434, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.1718, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.1094, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.5685, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.2856, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.7159, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.2823, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.1913, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.2656, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.2430, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.2192, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.3908, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.4586, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3868808567603748, 'recall': 0.3524390243902439, 'f1': 0.36885768985322265, 'number': 820}, 'P': {'precision': 0.47530864197530864, 'recall': 0.699205448354143, 'f1': 0.5659163987138263, 'number': 881}, 'overall_precision': 0.44297601566324035, 'overall_recall': 0.5320399764844209, 'overall_f1': 0.4834401709401709, 'overall_accuracy': 0.6690482038000402}
			------------EPOCH 20---------------
Loss:  tensor(18.2971, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.5113, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.8012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.4603, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.2322, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.3600, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.0457, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.9383, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.5932, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.5184, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.4829, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.2892, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.7728, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.7663, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.8897, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.4306, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(62.4897, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.3529, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.1891, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.1662, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.8511, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.8533, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.8874, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.9721, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.5666, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.1755, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.1596, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.6699, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.0575, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.1187, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.0418, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.8647, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.7138, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.4825, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.1047, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3944223107569721, 'recall': 0.3621951219512195, 'f1': 0.3776223776223776, 'number': 820}, 'P': {'precision': 0.4795050270688322, 'recall': 0.7037457434733257, 'f1': 0.5703771849126035, 'number': 881}, 'overall_precision': 0.44819159335288367, 'overall_recall': 0.5390946502057613, 'overall_f1': 0.4894582332532693, 'overall_accuracy': 0.6703878184375628}
			------------EPOCH 21---------------
Loss:  tensor(16.7410, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.7092, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.8960, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.1309, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.8869, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.2856, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.4079, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.4370, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.9071, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.8349, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.8889, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.4007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.1079, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.0100, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.1590, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.6248, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.2805, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.0341, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.4871, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.3815, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.5526, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.3400, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.7042, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.9882, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.6810, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.8632, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.8864, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.7861, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.4758, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.5580, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.6859, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.7609, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.8365, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.8899, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.1947, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3949468085106383, 'recall': 0.3621951219512195, 'f1': 0.37786259541984735, 'number': 820}, 'P': {'precision': 0.47809377401998465, 'recall': 0.7060158910329172, 'f1': 0.5701191567369386, 'number': 881}, 'overall_precision': 0.44763760350706283, 'overall_recall': 0.540270429159318, 'overall_f1': 0.4896110815130527, 'overall_accuracy': 0.6702538569738106}
			------------EPOCH 22---------------
Loss:  tensor(14.5945, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.0665, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.2193, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.6515, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.7697, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.3472, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.5063, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.5985, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.2469, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.0604, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.0607, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.8788, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.3668, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.7067, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.8662, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.6667, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.5052, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.6545, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.2479, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.0200, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.5810, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.8113, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.5282, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.2011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.3607, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.8399, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.1457, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.4576, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.9617, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.2116, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.1522, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.1822, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.7152, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.7053, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.8105, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.39300134589502017, 'recall': 0.35609756097560974, 'f1': 0.3736404350607805, 'number': 820}, 'P': {'precision': 0.4749807544264819, 'recall': 0.7003405221339387, 'f1': 0.5660550458715596, 'number': 881}, 'overall_precision': 0.44515181194906955, 'overall_recall': 0.5343915343915344, 'overall_f1': 0.48570665241784666, 'overall_accuracy': 0.6706780682756927}
			------------EPOCH 23---------------
Loss:  tensor(13.4601, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.2639, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.9373, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.3352, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.2560, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.6020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.7459, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.0386, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.8208, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.7669, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.6251, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.7964, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.9140, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.4735, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.4758, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.5531, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.9976, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.5749, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.2963, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.8410, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.8239, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.3006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.3294, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.6765, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.2670, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.0080, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.7638, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.3827, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.5433, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.0065, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.1798, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.5364, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.3779, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.7179, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.0204, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.39809782608695654, 'recall': 0.3573170731707317, 'f1': 0.37660668380462725, 'number': 820}, 'P': {'precision': 0.4760432766615147, 'recall': 0.699205448354143, 'f1': 0.5664367816091955, 'number': 881}, 'overall_precision': 0.44778325123152707, 'overall_recall': 0.5343915343915344, 'overall_f1': 0.48726882873224336, 'overall_accuracy': 0.6715934716113331}
			------------EPOCH 24---------------
Loss:  tensor(12.8022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.4410, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.8065, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.3947, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.3426, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.0744, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.1909, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.8467, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.6847, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.2062, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.6121, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.7948, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.6553, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.3215, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.3122, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.8368, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.0758, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.6349, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.5832, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.1601, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.2447, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.8711, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.1184, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.7562, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.4046, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.3692, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.5296, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.4203, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.2203, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.9783, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.1757, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.4272, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.2480, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.9832, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.5321, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.39295392953929537, 'recall': 0.35365853658536583, 'f1': 0.3722721437740693, 'number': 820}, 'P': {'precision': 0.4798136645962733, 'recall': 0.7014755959137344, 'f1': 0.5698478561549101, 'number': 881}, 'overall_precision': 0.4481737413622902, 'overall_recall': 0.5338036449147561, 'overall_f1': 0.48725516501207405, 'overall_accuracy': 0.6700975685994329}
			------------EPOCH 25---------------
Loss:  tensor(12.0070, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.5330, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.6025, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.6547, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.1755, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.1614, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.8897, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.2692, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.0867, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.8234, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.5803, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.5285, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.4157, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.6153, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.7066, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.7035, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.4747, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.9959, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.6912, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.4662, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.0129, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.5512, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.3062, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.4197, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.5750, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.4712, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.7161, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.0909, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.5340, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.0397, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.5423, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.7772, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.4004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.5353, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.5001, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3934640522875817, 'recall': 0.3670731707317073, 'f1': 0.3798107255520505, 'number': 820}, 'P': {'precision': 0.46273291925465837, 'recall': 0.6765039727582293, 'f1': 0.549562010142923, 'number': 881}, 'overall_precision': 0.4369215781782757, 'overall_recall': 0.527336860670194, 'overall_f1': 0.4778902503995738, 'overall_accuracy': 0.6677085891625176}
			------------EPOCH 26---------------
Loss:  tensor(10.5115, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.6365, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.0408, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.2312, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.9395, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.6154, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.4440, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.2496, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.2197, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.7651, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.0365, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.1907, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.4867, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.5179, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.6220, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.7846, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.9722, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.0058, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.2477, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.4352, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.6053, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.6007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.2660, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.1479, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.8950, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.8209, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.0215, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.2054, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.1100, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.7676, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.9282, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.8560, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.4403, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.0458, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.3529, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3780332056194125, 'recall': 0.36097560975609755, 'f1': 0.3693075483468496, 'number': 820}, 'P': {'precision': 0.45372460496614, 'recall': 0.684449489216799, 'f1': 0.5457013574660634, 'number': 881}, 'overall_precision': 0.4256628787878788, 'overall_recall': 0.5285126396237507, 'overall_f1': 0.4715447154471545, 'overall_accuracy': 0.6648060907812186}
			------------EPOCH 27---------------
Loss:  tensor(10.4447, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.6802, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.4083, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.1484, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.5090, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.3827, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.0682, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.1939, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.2317, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.8096, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.4238, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.3844, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.8302, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.5880, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.4571, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.7442, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.6907, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.1578, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.6996, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.9884, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.6027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.9616, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.9927, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.2493, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.7134, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.8461, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.9652, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.7539, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.7315, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.2609, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.4051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.0273, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.9186, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.6042, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.6568, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4013986013986014, 'recall': 0.35, 'f1': 0.37394136807817585, 'number': 820}, 'P': {'precision': 0.4692820133234641, 'recall': 0.7196367763904654, 'f1': 0.568100358422939, 'number': 881}, 'overall_precision': 0.4457889641819942, 'overall_recall': 0.5414462081128748, 'overall_f1': 0.48898327581629947, 'overall_accuracy': 0.6649623791555962}
			------------EPOCH 28---------------
Loss:  tensor(9.6602, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.5195, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.1035, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.4670, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.2944, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.5211, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.9530, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.0366, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.0395, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.0714, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.8561, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.1491, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.5333, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.5307, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.9808, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.2461, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.5062, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.1373, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.7127, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.6773, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.9441, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.8166, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.9357, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.2995, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.9648, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.0448, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.1637, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.2097, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.3276, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.7955, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.2088, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.0536, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.5459, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.0609, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.9493, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.40273972602739727, 'recall': 0.35853658536585364, 'f1': 0.37935483870967746, 'number': 820}, 'P': {'precision': 0.47228549734244496, 'recall': 0.7060158910329172, 'f1': 0.5659690627843494, 'number': 881}, 'overall_precision': 0.4474841231069858, 'overall_recall': 0.538506760728983, 'overall_f1': 0.4887940234791889, 'overall_accuracy': 0.6658108017593606}
			------------EPOCH 29---------------
Loss:  tensor(9.4631, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.9179, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.0499, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.8007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.3492, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.0443, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.5091, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.6974, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.0661, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.9114, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.5070, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.1928, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.3930, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.0709, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.7865, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.8093, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.9519, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.2732, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.8642, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.6297, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.7116, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.8326, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.7257, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.1694, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.1453, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.5393, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.7091, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.1651, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.2094, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.9653, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.7053, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.6571, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.0928, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.5715, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.6373, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4002659574468085, 'recall': 0.3670731707317073, 'f1': 0.38295165394402036, 'number': 820}, 'P': {'precision': 0.47090352220520676, 'recall': 0.6980703745743473, 'f1': 0.5624142661179699, 'number': 881}, 'overall_precision': 0.44509232264334303, 'overall_recall': 0.538506760728983, 'overall_f1': 0.48736366054801805, 'overall_accuracy': 0.6664136283462457}
			------------EPOCH 30---------------
Loss:  tensor(9.0561, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.3874, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.2442, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.4579, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.1760, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.6295, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.0537, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.7152, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.0266, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.9090, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.9142, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.8558, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.9221, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.8892, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.9379, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.1051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.8983, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.4267, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.4541, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.1758, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.7568, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.9286, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.1337, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.9713, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.8299, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.6795, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.9500, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.2497, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.7137, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.8844, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.8076, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.7044, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.3333, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.9107, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.0589, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.39648173207036536, 'recall': 0.3573170731707317, 'f1': 0.3758819756254009, 'number': 820}, 'P': {'precision': 0.46827794561933533, 'recall': 0.7037457434733257, 'f1': 0.562358276643991, 'number': 881}, 'overall_precision': 0.44255937954435287, 'overall_recall': 0.5367430922986478, 'overall_f1': 0.4851222104144527, 'overall_accuracy': 0.6657438210274844}


		-------------RUN 3-----------
			------------EPOCH 1---------------
Loss:  tensor(1641.6794, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2865.1206, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2538.2188, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2877.3135, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2357.3384, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1283.6992, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2182.4700, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2293.0142, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2035.9379, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2295.4216, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3651.8784, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1394.6355, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1543.8591, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2390.2246, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2358.1045, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1808.4834, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1932.8208, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1964.2513, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2373.2397, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1879.8766, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2417.6604, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1275.6577, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1261.5717, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3379.3523, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2568.5857, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2678.2329, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(916.7557, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1172.7854, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1206.5939, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1850.6985, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1922.9617, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1459.4790, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3156.5278, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2576.9673, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3219.0457, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1735.2355, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'number': 781}, 'P': {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'number': 867}, 'overall_precision': 0.0, 'overall_recall': 0.0, 'overall_f1': 0.0, 'overall_accuracy': 0.28806858775899025}
			------------EPOCH 2---------------
Loss:  tensor(1267.2026, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2223.0774, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2019.5601, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2134.2822, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2048.7925, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1054.0645, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1636.7004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1733.9874, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1582.5962, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1853.4596, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3034.1660, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1150.7762, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1339.7786, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2006.7526, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1968.4133, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1437.1316, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1493.8586, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1581.0972, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2110.0466, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1602.4384, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2102.3467, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1060.8539, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1092.7693, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3030.5796, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2277.6655, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2359.6536, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(806.9227, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1056.5439, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1086.6929, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1583.7869, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1704.7703, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1314.8330, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2719.3394, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2108.6001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2513.5920, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1422.1470, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.040755467196819085, 'recall': 0.052496798975672214, 'f1': 0.04588696138780078, 'number': 781}, 'P': {'precision': 0.13649025069637882, 'recall': 0.1695501730103806, 'f1': 0.15123456790123457, 'number': 867}, 'overall_precision': 0.09025444071051368, 'overall_recall': 0.11407766990291263, 'overall_f1': 0.10077727150897883, 'overall_accuracy': 0.524791617051679}
			------------EPOCH 3---------------
Loss:  tensor(1014.9267, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1784.1521, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1641.8813, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1521.0459, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1674.5017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(862.5020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1245.3420, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1280.8591, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1240.9128, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1373.5549, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2443.5994, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(924.4585, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1118.5128, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1818.9347, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1564.7424, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(992.8486, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(999.1180, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1227.5560, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1895.8562, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1441.5594, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1761.0684, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(778.1898, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(927.7854, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2498.5786, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1904.1019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2012.9409, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(686.3820, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(891.5276, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(943.1263, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1337.8676, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1382.7992, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1130.0320, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2242.0581, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1631.4475, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1803.6208, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1118.8062, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.08584686774941995, 'recall': 0.09475032010243278, 'f1': 0.0900791235544735, 'number': 781}, 'P': {'precision': 0.27662835249042145, 'recall': 0.41637831603229525, 'f1': 0.33241252302025776, 'number': 867}, 'overall_precision': 0.20073834794646978, 'overall_recall': 0.26395631067961167, 'overall_f1': 0.22804718217562256, 'overall_accuracy': 0.6000714455822815}
			------------EPOCH 4---------------
Loss:  tensor(825.5795, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1443.9291, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1248.4156, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(960.6661, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1326.8921, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(681.0671, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(965.0862, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(980.6677, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1007.8718, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1044.5576, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1911.3298, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(731.5806, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(896.9333, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1505.2468, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1143.3701, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(561.9956, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(699.4940, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(848.0142, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1492.6259, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1045.5684, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1338.7902, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(568.7583, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(831.0970, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1848.5229, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1381.3351, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1573.2463, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(509.7280, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(572.2043, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(713.3822, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1124.8411, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1063.6367, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(875.9603, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1793.3857, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1121.2524, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1196.0261, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(698.8904, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.17048054919908467, 'recall': 0.19078104993597952, 'f1': 0.1800604229607251, 'number': 781}, 'P': {'precision': 0.35824345146379044, 'recall': 0.5363321799307958, 'f1': 0.4295612009237875, 'number': 867}, 'overall_precision': 0.2826887661141805, 'overall_recall': 0.3725728155339806, 'overall_f1': 0.32146596858638743, 'overall_accuracy': 0.6249583234103357}
			------------EPOCH 5---------------
Loss:  tensor(577.5554, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1020.2346, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(857.4784, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(617.2208, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1072.0754, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(519.0947, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(702.7523, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(655.6592, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(736.6792, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(691.4535, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1325.7864, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(477.5035, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(584.2473, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1001.2297, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(806.8333, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(330.9152, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(423.6245, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(519.5641, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(878.6122, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(761.6617, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(897.0164, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(326.9554, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(555.0914, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(942.3560, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(808.6668, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(928.6691, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(319.1972, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(268.4398, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(440.0866, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(629.6262, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(789.0368, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(596.4232, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1121.7849, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(737.0704, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(648.9778, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(397.8167, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.2662337662337662, 'recall': 0.15749039692701663, 'f1': 0.1979082864038616, 'number': 781}, 'P': {'precision': 0.3893387314439946, 'recall': 0.6655132641291811, 'f1': 0.4912728820774799, 'number': 867}, 'overall_precision': 0.360082304526749, 'overall_recall': 0.42475728155339804, 'overall_f1': 0.3897550111358575, 'overall_accuracy': 0.6391045487020719}
			------------EPOCH 6---------------
Loss:  tensor(535.7324, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(863.7506, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(716.6899, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(383.2486, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(491.4446, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(293.9306, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(407.2286, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(356.1423, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(396.9118, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(365.3444, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(787.5902, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(307.8860, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(379.0806, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(477.7775, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(420.0824, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(208.5240, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(227.0765, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(320.5313, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(473.7040, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(571.0967, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(537.3153, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(188.9595, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(355.4693, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(474.7784, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(388.5188, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(472.1713, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(257.0274, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(192.0930, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(279.8719, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(477.4235, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(640.2545, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(424.0550, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(800.0922, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(503.1425, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(350.8988, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(206.0665, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.25793650793650796, 'recall': 0.08322663252240717, 'f1': 0.12584704743465633, 'number': 781}, 'P': {'precision': 0.3275862068965517, 'recall': 0.635524798154556, 'f1': 0.4323264025107886, 'number': 867}, 'overall_precision': 0.3185108583247156, 'overall_recall': 0.3737864077669903, 'overall_f1': 0.3439419318816303, 'overall_accuracy': 0.602691116932603}
			------------EPOCH 7---------------
Loss:  tensor(462.9568, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(519.4208, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(526.2981, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(361.7583, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(357.9653, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(275.5337, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(401.9664, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(260.2137, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(243.9647, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(414.2216, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(705.3772, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(169.9637, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(273.7411, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(272.1943, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(258.2495, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(148.7252, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(215.8098, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(332.3176, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(573.4917, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(403.4351, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(692.1733, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(348.8585, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(480.6726, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(611.0823, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(294.2307, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(379.1161, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(253.4704, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(200.6208, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(231.2610, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(311.8499, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(392.4981, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(290.9794, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(717.6550, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(437.7025, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(233.3914, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(261.3424, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.15403422982885084, 'recall': 0.08066581306017925, 'f1': 0.10588235294117646, 'number': 781}, 'P': {'precision': 0.28663967611336033, 'recall': 0.4083044982698962, 'f1': 0.3368220742150333, 'number': 867}, 'overall_precision': 0.25364963503649635, 'overall_recall': 0.2530339805825243, 'overall_f1': 0.2533414337788578, 'overall_accuracy': 0.5930459633246011}
			------------EPOCH 8---------------
Loss:  tensor(646.5689, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(777.6156, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(747.1572, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(678.7571, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(324.7770, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(106.0607, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(278.8922, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(231.1169, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(377.8826, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(294.3256, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(530.7957, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(195.2916, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(943.6360, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1309.1053, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(891.6812, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(386.4568, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(433.6861, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(447.5719, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(917.0779, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1165.0797, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(512.4497, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(233.3180, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(431.3889, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(797.3472, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(549.5280, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(910.7746, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(328.5924, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(297.9396, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(894.8508, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1755.6135, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1315.8345, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(969.4926, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1659.6449, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1629.5923, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2364.8604, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1569.3711, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.22773279352226722, 'recall': 0.2880921895006402, 'f1': 0.25438100621820237, 'number': 781}, 'P': {'precision': 0.401920438957476, 'recall': 0.33794694348327564, 'f1': 0.36716791979949875, 'number': 867}, 'overall_precision': 0.30168899242865466, 'overall_recall': 0.3143203883495146, 'overall_f1': 0.30787518573551265, 'overall_accuracy': 0.590092879256966}
			------------EPOCH 9---------------
Loss:  tensor(347.5723, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(359.1472, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(428.4157, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(409.7870, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(580.3495, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(220.5018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(454.4050, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(310.7395, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(721.6732, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(942.4576, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1597.5452, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(510.9563, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(684.9135, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1035.4783, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(751.6045, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(345.6609, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(623.1640, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(712.2183, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1114.1614, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(915.8203, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1020.9521, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(726.0410, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1389.0422, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2110.4775, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2083.7246, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1894.4938, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(378.2864, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(656.3163, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(588.2786, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1049.1158, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1258.6000, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1157.8594, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2036.1204, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1066.9608, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1147.5869, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(572.7783, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.15354980737479362, 'recall': 0.35723431498079383, 'f1': 0.21478060046189376, 'number': 781}, 'P': {'precision': 0.2642436149312377, 'recall': 0.3102652825836217, 'f1': 0.28541114058355443, 'number': 867}, 'overall_precision': 0.19329805996472663, 'overall_recall': 0.3325242718446602, 'overall_f1': 0.24447914343073832, 'overall_accuracy': 0.6103596094308169}
			------------EPOCH 10---------------
Loss:  tensor(531.0864, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(808.4772, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1043.6346, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(574.9017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1256.5148, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(594.3221, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(603.2488, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(517.8560, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(581.0123, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1094.8020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1749.7178, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(676.9156, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(699.3134, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1184.2338, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(992.1138, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(439.2206, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(398.9731, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(580.5555, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1044.4231, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(692.7470, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1189.6941, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(439.0180, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(550.0056, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1415.3519, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(938.1758, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1184.4653, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(401.0615, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(415.4357, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(449.3814, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(679.7754, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(698.8533, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(566.0785, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(974.2181, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(645.9888, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(821.0759, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(609.0783, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.1345083487940631, 'recall': 0.1856594110115237, 'f1': 0.15599784830554064, 'number': 781}, 'P': {'precision': 0.38883447600391774, 'recall': 0.45790080738177624, 'f1': 0.4205508474576271, 'number': 867}, 'overall_precision': 0.2582181991424488, 'overall_recall': 0.3288834951456311, 'overall_f1': 0.2892981051507873, 'overall_accuracy': 0.6432483924743987}
			------------EPOCH 11---------------
Loss:  tensor(353.5839, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(542.1574, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(491.6318, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(413.0582, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(492.9109, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(238.3781, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(437.7205, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(331.7724, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(387.0799, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(387.8224, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(833.1170, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(226.0248, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(469.1360, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(562.8364, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(388.8828, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(234.2698, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(315.6472, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(325.4458, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(513.8038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(471.9254, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(520.5837, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(255.5114, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(558.4902, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(543.5790, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(682.7365, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(671.7138, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(148.8776, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(165.0794, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(222.4827, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(422.5145, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(783.3155, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(452.5487, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(919.1088, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(439.8488, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(283.2067, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(284.0467, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.275, 'recall': 0.323943661971831, 'f1': 0.29747207524985303, 'number': 781}, 'P': {'precision': 0.46255144032921813, 'recall': 0.6482122260668973, 'f1': 0.5398655139289145, 'number': 867}, 'overall_precision': 0.38173302107728335, 'overall_recall': 0.4945388349514563, 'overall_f1': 0.43087496695744115, 'overall_accuracy': 0.6773517504167659}
			------------EPOCH 12---------------
Loss:  tensor(263.6436, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(309.5601, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(321.7712, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(191.8253, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(269.7260, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(111.3358, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(204.2614, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(150.1424, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(173.6409, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(287.4281, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(453.7484, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(148.0509, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(215.4994, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(254.4563, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(185.3571, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(140.4955, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(125.7527, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(139.4008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(272.9146, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(211.0480, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(291.2721, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(95.5736, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(283.4675, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(264.6793, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(236.0012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(190.7353, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(96.2683, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(90.2666, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(146.6073, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(177.3237, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(222.3744, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(154.5441, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(270.6270, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(237.0916, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(131.1236, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(107.7327, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.30689655172413793, 'recall': 0.4558258642765685, 'f1': 0.36682122617207624, 'number': 781}, 'P': {'precision': 0.4913978494623656, 'recall': 0.5271049596309112, 'f1': 0.5086254869226489, 'number': 867}, 'overall_precision': 0.3889952153110048, 'overall_recall': 0.4933252427184466, 'overall_f1': 0.434991974317817, 'overall_accuracy': 0.6757085020242914}
			------------EPOCH 13---------------
Loss:  tensor(186.4055, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(152.9268, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(246.6548, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(120.5257, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(193.4151, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(66.6192, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(162.7908, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(101.9274, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(116.6923, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(225.7945, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(319.1456, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(94.9670, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(166.7903, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(125.9297, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(111.9767, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(102.3805, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(95.3330, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(102.3686, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(186.8800, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(144.8240, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(210.3530, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(67.8467, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(178.5879, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(154.7507, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(159.9894, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(143.4210, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(59.1456, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(67.3900, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(83.7082, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(138.0357, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(155.8434, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(87.0220, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(186.4648, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(152.1206, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(62.8376, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(66.7976, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3458646616541353, 'recall': 0.353393085787452, 'f1': 0.34958834705509817, 'number': 781}, 'P': {'precision': 0.475, 'recall': 0.657439446366782, 'f1': 0.5515239477503628, 'number': 867}, 'overall_precision': 0.42342342342342343, 'overall_recall': 0.5133495145631068, 'overall_f1': 0.4640702139330774, 'overall_accuracy': 0.6827816146701595}
			------------EPOCH 14---------------
Loss:  tensor(120.5662, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(86.4788, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(101.0430, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(74.8859, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(109.7271, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.8838, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(72.6405, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.4901, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(73.8134, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(153.0726, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(168.0822, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.6407, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(127.8589, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(62.4068, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.0686, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(70.9985, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.0888, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(91.1454, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(130.0041, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(118.7702, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(136.2760, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.8191, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(120.2882, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(104.9231, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(78.5096, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(110.5251, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.5514, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(62.4287, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(80.1390, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(119.1201, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(98.1747, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(61.2454, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(142.2404, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(106.4294, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(59.0547, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(75.1790, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3900709219858156, 'recall': 0.4225352112676056, 'f1': 0.40565457897971724, 'number': 781}, 'P': {'precision': 0.4919064748201439, 'recall': 0.6309111880046137, 'f1': 0.5528044466902476, 'number': 867}, 'overall_precision': 0.44790602655771194, 'overall_recall': 0.5321601941747572, 'overall_f1': 0.4864115363283416, 'overall_accuracy': 0.6826149083115027}
			------------EPOCH 15---------------
Loss:  tensor(101.5225, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.9684, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(77.5200, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.2985, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(91.4367, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.3159, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.3195, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.1991, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.3901, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(116.9148, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(133.0401, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.0873, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(111.0151, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(59.9252, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.8508, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.7327, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.1261, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.9055, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(92.7071, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(77.1481, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(95.5074, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.3824, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(84.6553, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(75.2573, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.4491, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(87.5671, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.9493, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.1302, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.7729, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(67.5775, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(71.4748, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.2580, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(79.7038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(69.7997, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.3312, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.9539, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.41043890865954924, 'recall': 0.44302176696542894, 'f1': 0.4261083743842365, 'number': 781}, 'P': {'precision': 0.5008896797153025, 'recall': 0.649365628604383, 'f1': 0.5655449522852839, 'number': 867}, 'overall_precision': 0.4621250635485511, 'overall_recall': 0.5515776699029126, 'overall_f1': 0.5029045643153527, 'overall_accuracy': 0.6846630150035723}
			------------EPOCH 16---------------
Loss:  tensor(76.5623, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.4116, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.8887, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.8964, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.8132, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.7605, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.5831, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.9288, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.3114, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(97.7760, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(91.4638, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.1864, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(89.9547, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.8296, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.1851, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.6125, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.5376, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.2420, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(77.9654, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.2396, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(68.9630, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.1020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(65.2127, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.5798, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.5521, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(67.3824, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.5286, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.5427, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.5211, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.6158, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.2200, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.4103, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.2134, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.7290, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.9477, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.3925, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.41530740276035133, 'recall': 0.4238156209987196, 'f1': 0.4195183776932826, 'number': 781}, 'P': {'precision': 0.4901287553648069, 'recall': 0.6585928489042676, 'f1': 0.5620078740157481, 'number': 867}, 'overall_precision': 0.4597349643221203, 'overall_recall': 0.5473300970873787, 'overall_f1': 0.4997229916897507, 'overall_accuracy': 0.6842581567039772}
			------------EPOCH 17---------------
Loss:  tensor(58.2244, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.3151, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.3886, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.6444, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.2439, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.4159, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.0044, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.5969, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.3579, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(85.1925, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(72.4680, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.8060, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(71.5101, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.2383, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.8756, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.9762, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.6632, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.1416, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(62.8941, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.7058, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.1394, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.5502, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.5226, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.3052, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.0411, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.5226, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.1343, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.8144, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.7405, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.1964, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.3318, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.9473, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.3597, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.0575, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.8325, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.5725, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4099756690997567, 'recall': 0.43149807938540335, 'f1': 0.4204616344354336, 'number': 781}, 'P': {'precision': 0.5048715677590788, 'recall': 0.657439446366782, 'f1': 0.5711422845691383, 'number': 867}, 'overall_precision': 0.4648898001025115, 'overall_recall': 0.5503640776699029, 'overall_f1': 0.5040288969158099, 'overall_accuracy': 0.6844486782567278}
			------------EPOCH 18---------------
Loss:  tensor(43.8239, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.8562, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.4226, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.7312, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.4931, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.4109, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.1253, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.7301, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.4101, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(73.6495, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.2563, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.9695, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.2743, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.6353, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.5128, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.0244, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.6876, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.2832, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.6290, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.6532, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.4289, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.3541, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.5067, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.9497, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.8851, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.5000, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.9830, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.3808, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.6222, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.5131, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.6989, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.1098, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.1121, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.3664, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.2737, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.7690, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.41262135922330095, 'recall': 0.4353393085787452, 'f1': 0.4236760124610592, 'number': 781}, 'P': {'precision': 0.5, 'recall': 0.657439446366782, 'f1': 0.5680119581464873, 'number': 867}, 'overall_precision': 0.46334012219959264, 'overall_recall': 0.5521844660194175, 'overall_f1': 0.5038759689922481, 'overall_accuracy': 0.6827101690878781}
			------------EPOCH 19---------------
Loss:  tensor(27.3294, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.2824, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.7015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.9518, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.9839, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.7960, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.3971, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.9783, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.4003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.7884, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.4943, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.1942, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.6305, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.3974, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.9849, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.4395, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.9823, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.5023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.4642, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.0981, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.2824, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.2990, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.6364, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.5601, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.1638, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.2086, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.2988, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.8549, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.5223, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.0889, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.1542, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.4167, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.7637, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.6918, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.1811, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.3278, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4054726368159204, 'recall': 0.4174135723431498, 'f1': 0.4113564668769716, 'number': 781}, 'P': {'precision': 0.503056768558952, 'recall': 0.6643598615916955, 'f1': 0.5725646123260437, 'number': 867}, 'overall_precision': 0.4628014366341714, 'overall_recall': 0.5473300970873787, 'overall_f1': 0.5015290519877676, 'overall_accuracy': 0.6799476065729936}
			------------EPOCH 20---------------
Loss:  tensor(18.8713, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.2740, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.9960, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.2786, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.0382, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.6737, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.0392, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.6962, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.7357, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.8798, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.8074, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.1925, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.8543, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.1880, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.2493, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.6754, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.9656, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.8554, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.5611, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.6930, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.4417, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.8801, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.1872, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.6453, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.1177, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.2693, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.9135, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.8630, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.6723, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.3031, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.1151, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.6267, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.5624, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.4157, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.8343, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.1617, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.40648379052369077, 'recall': 0.4174135723431498, 'f1': 0.41187618445988633, 'number': 781}, 'P': {'precision': 0.5030621172353456, 'recall': 0.6632064590542099, 'f1': 0.572139303482587, 'number': 867}, 'overall_precision': 0.4632390745501285, 'overall_recall': 0.5467233009708737, 'overall_f1': 0.501530754244364, 'overall_accuracy': 0.6793522267206478}
			------------EPOCH 21---------------
Loss:  tensor(14.7935, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.2109, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.3711, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.3035, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.0398, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.0965, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.5714, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.0873, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.6416, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.2359, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.5561, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.7835, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.0585, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.5187, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.9916, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.5217, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.3279, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.3445, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.2817, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.6771, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.9215, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.7754, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.4139, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.3673, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.5334, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.9225, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.4975, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.3094, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.2286, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.5509, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.1759, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.4327, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.5837, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.9828, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.9425, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.2452, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4124223602484472, 'recall': 0.42509603072983354, 'f1': 0.41866330390920553, 'number': 781}, 'P': {'precision': 0.5039647577092511, 'recall': 0.6597462514417531, 'f1': 0.5714285714285714, 'number': 867}, 'overall_precision': 0.465979381443299, 'overall_recall': 0.5485436893203883, 'overall_f1': 0.503901895206243, 'overall_accuracy': 0.6797809002143368}
			------------EPOCH 22---------------
Loss:  tensor(12.1455, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.6287, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.6041, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.7517, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.6115, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.8308, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.5027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.7500, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.7886, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.9971, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.1948, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.7533, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.1599, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.0182, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.0212, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.7162, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.6847, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.2976, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.9003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.4193, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.8395, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.8927, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.0013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.4751, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.2312, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.9195, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.1233, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.2533, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.0016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.7787, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.8530, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.2400, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.4288, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.7830, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.4594, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.5545, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4114713216957606, 'recall': 0.4225352112676056, 'f1': 0.4169298799747315, 'number': 781}, 'P': {'precision': 0.500875656742557, 'recall': 0.6597462514417531, 'f1': 0.569437531110005, 'number': 867}, 'overall_precision': 0.46399176954732513, 'overall_recall': 0.5473300970873787, 'overall_f1': 0.502227171492205, 'overall_accuracy': 0.6770897832817337}
			------------EPOCH 23---------------
Loss:  tensor(10.4431, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.4091, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.5849, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.4332, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.5781, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.3267, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.8331, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.6309, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.1183, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.3741, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.3019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.9389, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.3788, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.7101, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.2648, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.1653, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.5082, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.7157, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.9336, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.5170, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.0909, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.1423, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.9818, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.8998, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.2604, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.0766, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.9618, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.5681, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.1340, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.1607, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.8955, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.2713, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.6103, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.9822, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.6190, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.0573, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.41635220125786165, 'recall': 0.4238156209987196, 'f1': 0.42005076142131986, 'number': 781}, 'P': {'precision': 0.505726872246696, 'recall': 0.6620530565167243, 'f1': 0.5734265734265734, 'number': 867}, 'overall_precision': 0.4689119170984456, 'overall_recall': 0.5491504854368932, 'overall_f1': 0.5058692006707657, 'overall_accuracy': 0.6773755656108598}
			------------EPOCH 24---------------
Loss:  tensor(9.1614, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.3868, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.0178, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.1685, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.8986, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.3350, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.4425, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.9408, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.8825, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.4056, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.0098, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.2838, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.2441, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.6090, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.7188, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.6752, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.8835, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.4125, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.4323, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.0436, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.1659, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.9186, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.6354, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.3039, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.5949, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.2982, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.2044, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.1636, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.8216, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.1971, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.9655, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.7863, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.7503, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.4110, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.7785, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.0178, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.41866330390920553, 'recall': 0.42509603072983354, 'f1': 0.42185514612452346, 'number': 781}, 'P': {'precision': 0.5065963060686016, 'recall': 0.6643598615916955, 'f1': 0.5748502994011976, 'number': 867}, 'overall_precision': 0.47046632124352333, 'overall_recall': 0.5509708737864077, 'overall_f1': 0.5075461151481275, 'overall_accuracy': 0.6777804239104549}
			------------EPOCH 25---------------
Loss:  tensor(8.0830, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.4758, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.8654, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.1428, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.4836, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.4959, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.2796, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.9653, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.8923, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.5386, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.3637, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.6063, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.7510, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.6206, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.1276, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.1571, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.4668, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.9345, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.1045, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.8441, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.8687, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.7411, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.6448, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.4915, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.8722, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.1088, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.5952, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.8409, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.3231, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.5385, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.3338, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.6999, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.3492, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.8993, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.6440, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.0813, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.432258064516129, 'recall': 0.42893725992317544, 'f1': 0.4305912596401028, 'number': 781}, 'P': {'precision': 0.5073465859982714, 'recall': 0.6770472895040369, 'f1': 0.5800395256916996, 'number': 867}, 'overall_precision': 0.4772256728778468, 'overall_recall': 0.5594660194174758, 'overall_f1': 0.5150837988826815, 'overall_accuracy': 0.6807573231721838}
			------------EPOCH 26---------------
Loss:  tensor(7.2115, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.6856, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.0750, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.4322, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.4220, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.5196, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.3328, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.1974, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.0951, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.4675, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.7571, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.0517, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.0308, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.6465, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.5047, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.8107, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.1530, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.5431, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.3243, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.6108, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.8113, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.3040, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.2807, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.9427, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.3508, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.1470, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.0206, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.7562, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.4715, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.5229, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.7581, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.6625, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.4181, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.8168, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.8769, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.3803, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.43918053777208704, 'recall': 0.43918053777208704, 'f1': 0.43918053777208704, 'number': 781}, 'P': {'precision': 0.5078397212543554, 'recall': 0.6724336793540946, 'f1': 0.5786600496277915, 'number': 867}, 'overall_precision': 0.4800414722654225, 'overall_recall': 0.5618932038834952, 'overall_f1': 0.5177523064020129, 'overall_accuracy': 0.6817575613241248}
			------------EPOCH 27---------------
Loss:  tensor(6.5280, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.0717, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.4969, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.8690, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.6070, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.9200, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.4788, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.5714, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.4546, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.9440, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.4418, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.6445, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.7640, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.6907, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.9682, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.5736, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.9117, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.1544, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.8606, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.6769, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.0775, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.9126, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.1228, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.1460, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.9106, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.1650, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.9178, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.8856, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.7898, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.5045, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.4595, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.4001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.6354, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.3481, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.1686, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.5181, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.43918053777208704, 'recall': 0.43918053777208704, 'f1': 0.43918053777208704, 'number': 781}, 'P': {'precision': 0.509170305676856, 'recall': 0.6724336793540946, 'f1': 0.5795228628230616, 'number': 867}, 'overall_precision': 0.48078920041536866, 'overall_recall': 0.5618932038834952, 'overall_f1': 0.5181869054280918, 'overall_accuracy': 0.6814717789949988}
			------------EPOCH 28---------------
Loss:  tensor(5.9418, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.5504, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.8681, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.2700, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.9023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.1391, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.7736, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.0169, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.8739, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.7660, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.4509, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.3079, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.6960, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.7558, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.5654, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.4458, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.7275, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.3166, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.0776, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.8652, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.5638, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.4877, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.1057, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.3122, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.0279, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.2726, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.9216, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.4199, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.0783, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.6102, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.4345, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.9288, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.1974, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.4231, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.7022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.7900, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4266666666666667, 'recall': 0.40973111395646605, 'f1': 0.41802743305029394, 'number': 781}, 'P': {'precision': 0.493573264781491, 'recall': 0.6643598615916955, 'f1': 0.5663716814159292, 'number': 867}, 'overall_precision': 0.46739697443922795, 'overall_recall': 0.5436893203883495, 'overall_f1': 0.5026647966339411, 'overall_accuracy': 0.6819480828768755}
			------------EPOCH 29---------------
Loss:  tensor(5.4453, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.1080, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.1809, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.6830, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.5713, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.8766, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.8781, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.6286, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.5708, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.2654, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.0857, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.9882, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.7567, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.1754, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.2646, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.6742, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.3911, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.8592, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.0231, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.3515, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.8944, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.9282, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.3874, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.8368, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.0611, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.5162, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.2124, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.6623, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.9008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.8336, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.7287, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.5991, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.5314, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.9333, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.3945, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.5056, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.43843031123139375, 'recall': 0.4148527528809219, 'f1': 0.42631578947368415, 'number': 781}, 'P': {'precision': 0.49700085689802914, 'recall': 0.6689734717416378, 'f1': 0.5703048180924287, 'number': 867}, 'overall_precision': 0.47429171038824763, 'overall_recall': 0.5485436893203883, 'overall_f1': 0.5087225661226787, 'overall_accuracy': 0.6831150273874732}
			------------EPOCH 30---------------
Loss:  tensor(5.0619, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.7132, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.6089, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.5379, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.0865, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.1533, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.3664, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.1688, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.0830, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.6096, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.4971, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.6598, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.0855, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.9962, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.5093, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.3466, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.4644, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.6279, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.2797, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.8438, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.7357, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.1849, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.2505, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.0771, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.1058, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.9723, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.4594, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.9968, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.3252, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.7734, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.6376, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.4407, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.3928, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.6030, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.6993, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.7287, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4218928164196123, 'recall': 0.4737516005121639, 'f1': 0.44632086851628466, 'number': 781}, 'P': {'precision': 0.5177238805970149, 'recall': 0.6401384083044983, 'f1': 0.5724600309437855, 'number': 867}, 'overall_precision': 0.4746023601847101, 'overall_recall': 0.5612864077669902, 'overall_f1': 0.5143174867945509, 'overall_accuracy': 0.6810907358894975}


		-------------RUN 4-----------
			------------EPOCH 1---------------
Loss:  tensor(2185.7854, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2615.5623, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1550.5710, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1938.2417, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2122.3486, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2849.6289, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3345.4644, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2700.6829, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3054.5688, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2420.8662, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2259.4941, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1602.2299, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1426.5570, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(992.2267, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2167.2227, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2514.1323, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1250.4891, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1523.1125, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2101.2290, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3198.2964, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1649.6527, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1996.5349, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1637.3730, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2640.6704, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2397.8923, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3005.9399, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1263.6011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1560.7350, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4036.3281, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2211.3811, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1526.2939, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2065.5808, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2386.2673, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1261.6560, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(918.2111, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1044.4376, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(903.7277, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.023622047244094488, 'recall': 0.004424778761061947, 'f1': 0.007453416149068323, 'number': 678}, 'P': {'precision': 0.07869142351900972, 'recall': 0.1100123609394314, 'f1': 0.09175257731958764, 'number': 809}, 'overall_precision': 0.07313195548489666, 'overall_recall': 0.06186953597848016, 'overall_f1': 0.06703096539162114, 'overall_accuracy': 0.5159479318600875}
			------------EPOCH 2---------------
Loss:  tensor(1594.8638, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2021.7213, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1038.6498, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1234.2374, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1291.1743, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1956.0492, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2784.7793, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2132.5522, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2341.4766, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1951.4396, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1752.9592, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1250.6453, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1128.1741, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(820.2704, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1781.7675, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2101.8672, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1026.8558, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1200.9792, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1536.8108, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2577.4460, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1290.0737, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1647.3694, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1352.2133, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2194.1870, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2113.9172, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2507.0391, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1074.5458, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1342.1216, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3422.1575, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1792.8354, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1256.7178, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1769.4965, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1956.4694, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1022.1023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(773.7509, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(879.5107, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(736.1492, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.11531190926275993, 'recall': 0.08997050147492626, 'f1': 0.10107705053852527, 'number': 678}, 'P': {'precision': 0.196875, 'recall': 0.23362175525339926, 'f1': 0.21368004522329, 'number': 809}, 'overall_precision': 0.16789791806581597, 'overall_recall': 0.16812373907195696, 'overall_f1': 0.16801075268817203, 'overall_accuracy': 0.5832622395905}
			------------EPOCH 3---------------
Loss:  tensor(1336.7039, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1819.8965, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(817.4084, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1013.2111, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(993.1554, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1606.0076, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2602.9980, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1940.4399, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2043.4447, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1805.3325, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1490.9980, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1042.0543, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(902.6168, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(670.1815, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1491.4399, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1757.4584, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(861.3937, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(976.7042, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1235.0200, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2133.8357, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1025.6154, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1399.8932, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1191.1307, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1875.4924, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1894.8530, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2046.9615, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(927.4923, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1149.1814, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3100.2349, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1496.3947, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1103.3958, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1514.6715, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1516.3792, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(809.3502, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(649.2987, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(739.5308, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(644.2435, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.15130674002751032, 'recall': 0.16224188790560473, 'f1': 0.15658362989323846, 'number': 678}, 'P': {'precision': 0.30033745781777277, 'recall': 0.3300370828182942, 'f1': 0.3144876325088339, 'number': 809}, 'overall_precision': 0.2332920792079208, 'overall_recall': 0.2535305985205111, 'overall_f1': 0.24299065420560748, 'overall_accuracy': 0.5994991331150067}
			------------EPOCH 4---------------
Loss:  tensor(1069.2113, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1524.0922, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(664.8062, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(782.7892, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(738.3671, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1306.8311, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2042.4706, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1602.4966, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1746.3118, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1504.3712, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1273.4894, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(867.2195, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(781.2615, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(583.8698, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1325.5195, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1470.5863, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(709.5635, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(786.1879, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(992.8830, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1686.4271, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(662.2980, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1085.8948, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(938.9497, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1474.4673, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1553.2200, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1559.9875, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(708.4998, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(921.4922, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2496.9216, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1138.0876, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(870.2245, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1213.7017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1076.4172, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(560.4627, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(495.2189, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(534.2993, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(604.4517, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.2515151515151515, 'recall': 0.1224188790560472, 'f1': 0.16468253968253968, 'number': 678}, 'P': {'precision': 0.38916666666666666, 'recall': 0.5772558714462299, 'f1': 0.4649079143852663, 'number': 809}, 'overall_precision': 0.35947712418300654, 'overall_recall': 0.3698722259583053, 'overall_f1': 0.3646005966191581, 'overall_accuracy': 0.6041225197457137}
			------------EPOCH 5---------------
Loss:  tensor(869.2383, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1117.0594, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(597.3246, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(488.5787, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(504.9886, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(954.6097, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1407.1167, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1281.5383, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1320.0227, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1051.0829, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1001.0438, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(637.8332, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(680.0466, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(432.1838, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1029.4751, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1244.3912, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(541.9038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(541.5443, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(774.8424, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1239.0054, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(422.9442, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(880.0695, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(644.8965, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1128.6240, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1173.2124, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1087.9243, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(521.1204, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(675.6559, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1725.0798, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(709.2189, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(623.4921, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(806.1754, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(830.8088, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(397.2692, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(351.0174, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(385.1546, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(541.6170, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.2913978494623656, 'recall': 0.3997050147492625, 'f1': 0.33706467661691536, 'number': 678}, 'P': {'precision': 0.44803370786516855, 'recall': 0.3943139678615575, 'f1': 0.41946088099934253, 'number': 809}, 'overall_precision': 0.35931790499390986, 'overall_recall': 0.39677202420981844, 'overall_f1': 0.3771172898689677, 'overall_accuracy': 0.629743787324215}
			------------EPOCH 6---------------
Loss:  tensor(627.9435, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(853.5964, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(431.4135, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(351.4585, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(360.9398, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(652.2253, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(928.7090, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(848.5907, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1081.0658, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(800.6429, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(841.0382, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(519.7681, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(593.7880, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(426.2908, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(887.6181, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(829.6749, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(335.8209, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(378.1703, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(458.0776, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(813.9169, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(278.7455, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(731.2169, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(593.2459, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1010.9886, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1001.0880, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(832.2659, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(427.4277, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(497.5047, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1316.8109, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(448.6927, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(468.1050, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(583.5155, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(628.1125, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(295.7510, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(275.0496, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(259.0689, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(440.7518, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.39740820734341253, 'recall': 0.2713864306784661, 'f1': 0.32252410166520595, 'number': 678}, 'P': {'precision': 0.42940199335548174, 'recall': 0.6390605686032138, 'f1': 0.5136612021857923, 'number': 809}, 'overall_precision': 0.4205158968206359, 'overall_recall': 0.4714189643577673, 'overall_f1': 0.4445149017121116, 'overall_accuracy': 0.6572089055232958}
			------------EPOCH 7---------------
Loss:  tensor(443.5287, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(561.8636, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(320.1507, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(195.4441, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(261.3419, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(410.6969, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(516.5987, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(508.2443, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(678.3360, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(492.2678, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(557.1264, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(310.9578, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(320.0839, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(194.9089, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(550.9021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(460.1611, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(198.9593, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(257.8583, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(335.9815, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(603.6749, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(226.3022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(591.8510, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(288.3202, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(632.1932, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(532.9427, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(540.6211, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(244.1967, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(293.7435, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(818.2841, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(267.3350, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(270.1430, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(352.2538, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(269.9657, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(159.5831, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(170.4927, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(180.2395, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(311.4071, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3460207612456747, 'recall': 0.2949852507374631, 'f1': 0.3184713375796179, 'number': 678}, 'P': {'precision': 0.4558258642765685, 'recall': 0.4400494437577256, 'f1': 0.4477987421383648, 'number': 809}, 'overall_precision': 0.4091243561442237, 'overall_recall': 0.37390719569603226, 'overall_f1': 0.39072382290934643, 'overall_accuracy': 0.6408619313647247}
			------------EPOCH 8---------------
Loss:  tensor(377.3627, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(601.1072, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(323.5975, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(156.5339, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(175.4736, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(240.7563, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(315.9752, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(259.9812, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(425.4915, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(315.4776, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(353.1109, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(209.1476, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(259.4109, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(161.5918, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(431.9890, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(248.4661, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(189.2993, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(290.7106, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(307.6259, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(507.3412, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(170.5689, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(351.6026, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(194.4570, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(525.9616, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(355.5930, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(467.6190, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(152.3368, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(210.8897, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(587.9827, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(179.1876, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(227.2063, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(272.0511, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(189.2062, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(124.3746, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(100.8823, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(117.9676, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(216.8641, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.24721603563474387, 'recall': 0.16371681415929204, 'f1': 0.19698314108252, 'number': 678}, 'P': {'precision': 0.35456110154905335, 'recall': 0.2546353522867738, 'f1': 0.29640287769784174, 'number': 809}, 'overall_precision': 0.30776699029126214, 'overall_recall': 0.21318090114324142, 'overall_f1': 0.25188716726261423, 'overall_accuracy': 0.5671079065415416}
			------------EPOCH 9---------------
Loss:  tensor(352.8748, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(382.6137, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(184.5117, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(162.1750, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(491.7112, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(370.2036, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(509.3965, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(400.1056, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(370.8880, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(356.0402, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(314.5966, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(219.2042, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(175.3303, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(88.4596, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(230.2379, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(170.1585, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(111.8013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(182.8058, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(235.1682, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(382.2906, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(201.1581, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(289.5073, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(212.7737, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(672.5634, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(616.4630, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(640.9940, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(171.3730, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(418.2992, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1360.9880, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(346.2584, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(382.3383, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(848.9232, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(313.6420, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(330.1661, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(220.0381, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(197.9129, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(161.1588, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.36054421768707484, 'recall': 0.5471976401179941, 'f1': 0.434680726420621, 'number': 678}, 'P': {'precision': 0.505420054200542, 'recall': 0.4610630407911001, 'f1': 0.482223658694247, 'number': 809}, 'overall_precision': 0.42105263157894735, 'overall_recall': 0.5003362474781439, 'overall_f1': 0.4572833435771358, 'overall_accuracy': 0.6775738228252195}
			------------EPOCH 10---------------
Loss:  tensor(245.8466, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(448.3752, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(157.8522, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(133.4267, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(151.1227, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(204.3722, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(285.7003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(226.0327, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(363.6607, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(272.2225, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(395.0088, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(276.0001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(341.5904, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(215.2291, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(554.8538, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(338.7771, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(239.3193, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(425.2885, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(830.6267, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1095.4956, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(169.0516, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(409.1265, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(293.7453, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(404.2890, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(304.0663, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(440.1034, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(174.3736, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(251.2392, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(469.9315, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(176.6884, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(151.2562, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(218.2523, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(160.6585, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(134.5755, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(93.1090, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(111.8303, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(140.7657, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3595238095238095, 'recall': 0.44542772861356933, 'f1': 0.39789196310935443, 'number': 678}, 'P': {'precision': 0.4548022598870056, 'recall': 0.5970333745364648, 'f1': 0.5163014430785676, 'number': 809}, 'overall_precision': 0.4127234490010515, 'overall_recall': 0.5279085406859448, 'overall_f1': 0.46326349955739154, 'overall_accuracy': 0.6508517489060737}
			------------EPOCH 11---------------
Loss:  tensor(377.6954, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(794.3821, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(187.2576, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(198.9082, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(228.8283, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(329.1800, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(673.0342, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(537.5009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(560.7970, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(618.3840, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(361.8780, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(250.7070, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(169.4906, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(124.3393, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(277.1142, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(256.7426, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(83.4481, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(124.8927, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(333.7371, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(375.5716, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(88.7506, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(175.7431, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(108.5917, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(260.3226, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(214.9404, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(370.4066, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(148.9815, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(179.5888, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(524.5233, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(133.5829, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(244.6181, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(242.5618, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(144.7188, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(106.1519, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(82.1777, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(86.7712, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(101.8630, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4104938271604938, 'recall': 0.39233038348082594, 'f1': 0.40120663650075417, 'number': 678}, 'P': {'precision': 0.48705882352941177, 'recall': 0.511742892459827, 'f1': 0.4990958408679928, 'number': 809}, 'overall_precision': 0.4539385847797063, 'overall_recall': 0.45729657027572296, 'overall_f1': 0.45561139028475717, 'overall_accuracy': 0.6788672702754768}
			------------EPOCH 12---------------
Loss:  tensor(154.6335, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(180.0553, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(97.7069, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.9456, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(92.5452, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(133.0644, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(161.2564, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(103.7654, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(190.5841, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(130.3293, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(159.9468, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(112.8253, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(145.5149, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.9333, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(168.0918, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(141.2363, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(96.5775, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(116.8149, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(163.6660, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(275.9070, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(72.1863, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(175.3037, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(101.3407, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(197.3603, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(186.6261, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(283.3568, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(96.5530, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(112.6582, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(256.1939, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(64.8774, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(103.3004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(123.6018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(96.5680, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.6418, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.2225, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(71.1686, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(107.4729, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.45695364238410596, 'recall': 0.20353982300884957, 'f1': 0.2816326530612245, 'number': 678}, 'P': {'precision': 0.4013206162876009, 'recall': 0.6761433868974042, 'f1': 0.503683241252302, 'number': 809}, 'overall_precision': 0.4114114114114114, 'overall_recall': 0.4606590450571621, 'overall_f1': 0.43464467005076146, 'overall_accuracy': 0.6668960013209676}
			------------EPOCH 13---------------
Loss:  tensor(252.0067, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(230.3031, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(205.7699, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(86.7799, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(112.5818, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(169.5742, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(240.0537, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(405.6766, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(177.5450, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(98.5847, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(145.5428, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(115.7559, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(79.8964, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.2344, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(88.4658, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(81.8493, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(100.2350, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(96.7515, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(114.8070, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(197.0604, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(115.7826, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(399.3507, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(314.5309, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(407.3241, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(217.7071, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(343.4020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(213.8559, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(181.1219, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(206.2145, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(59.8587, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(93.1637, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(115.6314, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(65.0458, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.4590, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.3786, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.4626, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(61.8211, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.39768339768339767, 'recall': 0.15191740412979352, 'f1': 0.21985058697972254, 'number': 678}, 'P': {'precision': 0.4400887573964497, 'recall': 0.7354758961681088, 'f1': 0.5506709856547894, 'number': 809}, 'overall_precision': 0.43327126008690253, 'overall_recall': 0.4694014794889038, 'overall_f1': 0.4506132989025178, 'overall_accuracy': 0.6596031593141977}
			------------EPOCH 14---------------
Loss:  tensor(121.3542, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(190.8181, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(84.4526, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.9487, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(112.3094, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(126.0203, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(226.9029, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(188.3687, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(766.5956, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(388.2678, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(994.8337, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(457.2542, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(181.1723, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(88.6551, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(210.2410, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(136.5792, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.6311, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(86.8048, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(134.3149, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(159.1797, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.1618, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(112.5532, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(82.1594, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(105.7832, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(177.6218, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(289.0453, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(120.0671, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(136.5657, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1455.3677, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(264.6108, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(383.6610, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1406.8313, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(154.0913, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(74.9480, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(103.6100, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(138.7777, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(100.0113, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4243759177679883, 'recall': 0.42625368731563423, 'f1': 0.42531272994849156, 'number': 678}, 'P': {'precision': 0.5370558375634518, 'recall': 0.65389369592089, 'f1': 0.5897435897435898, 'number': 809}, 'overall_precision': 0.4909963985594238, 'overall_recall': 0.5501008742434432, 'overall_f1': 0.5188709165873772, 'overall_accuracy': 0.6970030547375953}
			------------EPOCH 15---------------
Loss:  tensor(104.1151, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(125.2260, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(81.2162, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.2156, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.5650, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(97.9629, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(130.8781, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.4856, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(141.7740, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(84.6953, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(126.5925, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(72.9267, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(126.8800, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(198.1961, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(237.0420, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(279.2760, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(90.9766, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(335.7502, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(198.8972, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(495.3390, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(124.8366, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(548.0875, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(680.8286, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(615.4907, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(337.2973, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(279.5430, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(162.0757, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(265.3708, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(468.0856, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(136.5343, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(188.8250, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(154.4443, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(80.6213, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.8719, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.6739, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.1315, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(127.7531, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3875, 'recall': 0.6858407079646017, 'f1': 0.49520766773162933, 'number': 678}, 'P': {'precision': 0.5931558935361216, 'recall': 0.3856613102595797, 'f1': 0.4674157303370786, 'number': 809}, 'overall_precision': 0.4501738122827346, 'overall_recall': 0.5225285810356423, 'overall_f1': 0.4836601307189542, 'overall_accuracy': 0.6767206979112199}
			------------EPOCH 16---------------
Loss:  tensor(200.4397, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(185.3414, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(121.9573, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(156.8555, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(260.0628, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(373.2800, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(782.8771, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(546.8768, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(485.7965, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(352.5700, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(278.7932, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(251.1272, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(107.0737, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.9349, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(174.3900, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(166.2330, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.5320, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(69.3049, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(103.4844, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(148.7693, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.4958, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(96.7802, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(81.3929, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(82.3616, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(100.5555, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(204.0227, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(71.0174, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(71.7075, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(334.7461, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(93.9266, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(105.8590, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(127.4025, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(93.8436, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(80.9312, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(61.4243, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.2385, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(168.6109, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5089605734767025, 'recall': 0.20943952802359883, 'f1': 0.296760710553814, 'number': 678}, 'P': {'precision': 0.4869029275808937, 'recall': 0.7812113720642769, 'f1': 0.599905078310394, 'number': 809}, 'overall_precision': 0.49080532656943565, 'overall_recall': 0.5205110961667787, 'overall_f1': 0.5052219321148824, 'overall_accuracy': 0.676968379337865}
			------------EPOCH 17---------------
Loss:  tensor(105.9721, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(170.2841, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.3825, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.7676, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.0144, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(92.2906, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(195.4107, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(125.1145, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(123.6149, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(74.8042, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(105.1758, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(83.8877, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(116.2243, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(65.2591, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(110.6480, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(102.8200, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.3241, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.8835, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(86.7954, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(150.8375, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.2566, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(86.5620, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(71.8499, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(70.8065, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(81.9445, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(171.0889, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(68.2444, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.3240, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(133.2592, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.4101, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.0146, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(81.1476, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.7896, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.6009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.4321, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.3253, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.5780, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4453125, 'recall': 0.42035398230088494, 'f1': 0.43247344461305004, 'number': 678}, 'P': {'precision': 0.5417085427135678, 'recall': 0.6662546353522868, 'f1': 0.5975609756097561, 'number': 809}, 'overall_precision': 0.5039755351681957, 'overall_recall': 0.5541358439811701, 'overall_f1': 0.5278667520819986, 'overall_accuracy': 0.7056168643531386}
			------------EPOCH 18---------------
Loss:  tensor(57.7216, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(68.4049, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.0574, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.9271, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.9019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(68.0936, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(79.0255, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.9721, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(89.4326, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.1909, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(84.8027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.0486, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.5012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.5814, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.7875, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(61.0507, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.2453, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.2152, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(59.1087, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(89.4706, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.9298, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(82.9555, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(62.6590, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(64.4679, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(62.9253, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(149.3675, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.1954, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.4175, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(92.3099, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.9015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.0136, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(62.6055, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.3281, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.2756, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.5020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.1912, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.5033, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.40417690417690416, 'recall': 0.48525073746312686, 'f1': 0.4410187667560322, 'number': 678}, 'P': {'precision': 0.551487414187643, 'recall': 0.595797280593325, 'f1': 0.5727866904337492, 'number': 809}, 'overall_precision': 0.48045023696682465, 'overall_recall': 0.5453934095494284, 'overall_f1': 0.5108661417322835, 'overall_accuracy': 0.7060021465723643}
			------------EPOCH 19---------------
Loss:  tensor(43.0909, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(59.7415, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.2861, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.9763, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.4065, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.6184, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(66.1864, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.2041, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(81.1850, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.3782, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(68.6343, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.0615, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.1429, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.8894, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.4452, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.3271, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.2534, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.4284, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.0525, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(68.6816, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.1170, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(65.9887, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.4635, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.9058, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.4385, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(134.8641, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.8984, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.7440, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(69.0400, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.9153, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.6093, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.4837, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.0480, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.6951, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.8566, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.3248, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.7214, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.43784530386740333, 'recall': 0.46755162241887904, 'f1': 0.4522111269614836, 'number': 678}, 'P': {'precision': 0.5511302475780409, 'recall': 0.6328800988875154, 'f1': 0.5891829689298044, 'number': 809}, 'overall_precision': 0.5015124016938899, 'overall_recall': 0.5574983187626092, 'overall_f1': 0.5280254777070064, 'overall_accuracy': 0.7028648485015274}
			------------EPOCH 20---------------
Loss:  tensor(35.2920, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.3545, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.8713, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.7337, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.3670, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.7789, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.7694, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.8505, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(75.1187, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.7532, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.9945, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.5739, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.1688, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.7958, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.1598, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.2443, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.6515, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.6861, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.4869, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(59.7942, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.6558, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.6030, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.7842, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.8100, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.1206, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(124.2286, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.9588, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.8674, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.3585, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.9951, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.2538, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.6641, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.2910, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.7241, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.9603, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.8511, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.8394, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4357864357864358, 'recall': 0.44542772861356933, 'f1': 0.44055433989788473, 'number': 678}, 'P': {'precision': 0.5658031088082901, 'recall': 0.6749072929542645, 'f1': 0.6155580608793686, 'number': 809}, 'overall_precision': 0.51145958986731, 'overall_recall': 0.570275722932078, 'overall_f1': 0.5392686804451511, 'overall_accuracy': 0.7062773481575254}
			------------EPOCH 21---------------
Loss:  tensor(30.9530, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.3449, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.3553, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.3136, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.9662, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.4994, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.3483, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.3795, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(68.7867, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.5707, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.9297, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.9514, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.2303, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.7480, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.4267, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.8854, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.5361, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.0092, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.7545, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.3054, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.5290, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.8444, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.8132, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.3074, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.0380, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(113.1897, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.2726, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.6058, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.7882, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.3953, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.6139, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.7012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.7678, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.5659, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.6228, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.3878, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.1107, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.43045387994143486, 'recall': 0.4336283185840708, 'f1': 0.43203526818515803, 'number': 678}, 'P': {'precision': 0.557711950970378, 'recall': 0.6749072929542645, 'f1': 0.6107382550335572, 'number': 809}, 'overall_precision': 0.5054151624548736, 'overall_recall': 0.5648957632817754, 'overall_f1': 0.5335026992696094, 'overall_accuracy': 0.7040757354762364}
			------------EPOCH 22---------------
Loss:  tensor(31.7388, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.2745, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.7205, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.9229, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.0624, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.8358, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.4871, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.4979, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(65.5380, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.9240, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.6546, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.6566, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.2493, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.7349, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.6565, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.8408, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.3809, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.0963, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.4379, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.1641, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.3250, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.3495, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.6602, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.9661, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.8243, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(102.0754, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.8848, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.3204, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.7947, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.8896, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.4021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.8353, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.3376, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.1272, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.5974, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.4896, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.6579, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4185022026431718, 'recall': 0.42035398230088494, 'f1': 0.41942604856512133, 'number': 678}, 'P': {'precision': 0.5479591836734694, 'recall': 0.6637824474660075, 'f1': 0.6003353828954723, 'number': 809}, 'overall_precision': 0.4948826008428657, 'overall_recall': 0.5527908540685945, 'overall_f1': 0.5222363405336722, 'overall_accuracy': 0.6978286594930787}
			------------EPOCH 23---------------
Loss:  tensor(28.6529, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.4773, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.9043, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.2453, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.7317, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.8337, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.9626, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.7838, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.1876, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.3874, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.4056, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.8971, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.8217, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.5172, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.1342, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.6935, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.4526, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.0199, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.0113, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.1188, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.9050, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.8827, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.1217, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.8081, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.0131, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(88.8360, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.0721, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.2057, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.0733, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.2421, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.9392, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.5747, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.7647, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.9489, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.1874, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.8459, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.2383, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.41935483870967744, 'recall': 0.4218289085545723, 'f1': 0.42058823529411765, 'number': 678}, 'P': {'precision': 0.5539714867617108, 'recall': 0.6724351050679852, 'f1': 0.6074818537130096, 'number': 809}, 'overall_precision': 0.4987980769230769, 'overall_recall': 0.5581708137188971, 'overall_f1': 0.5268168835290383, 'overall_accuracy': 0.6951592041170157}
			------------EPOCH 24---------------
Loss:  tensor(24.6743, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.7723, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.0408, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.6162, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.1551, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.2835, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.3217, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.7291, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.7904, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.7847, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.3473, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.3185, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.5892, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.7419, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.1490, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.9948, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.5088, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.3879, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.7078, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.7970, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.6965, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.9631, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.4341, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.4685, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.8600, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(72.1219, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.7094, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.7338, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.6491, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.3456, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.2789, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.3975, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.9205, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.3467, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.7803, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.1611, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.8415, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4067055393586006, 'recall': 0.41150442477876104, 'f1': 0.40909090909090906, 'number': 678}, 'P': {'precision': 0.5537525354969574, 'recall': 0.6749072929542645, 'f1': 0.6083565459610029, 'number': 809}, 'overall_precision': 0.4934210526315789, 'overall_recall': 0.554808338937458, 'overall_f1': 0.5223171889838556, 'overall_accuracy': 0.693507994606049}
			------------EPOCH 25---------------
Loss:  tensor(20.2021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.3644, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.6574, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.8549, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.7477, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.1751, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.6816, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.1193, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.0580, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.8137, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.7200, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.1474, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.9702, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.4997, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.0731, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.0431, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.1219, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.9475, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.8470, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.2401, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.7743, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.1745, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.0725, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.8512, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.0090, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(65.7984, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.1491, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.0720, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.9508, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.0059, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.5360, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.3725, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.1555, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.5875, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.1959, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.7903, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.6215, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4370257966616085, 'recall': 0.4247787610619469, 'f1': 0.43081525804038895, 'number': 678}, 'P': {'precision': 0.5515151515151515, 'recall': 0.6749072929542645, 'f1': 0.6070038910505837, 'number': 809}, 'overall_precision': 0.5057610673135233, 'overall_recall': 0.5608607935440484, 'overall_f1': 0.5318877551020408, 'overall_accuracy': 0.6985992239315298}
			------------EPOCH 26---------------
Loss:  tensor(29.5872, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.1014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.5354, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.1192, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.3910, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.3420, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.7212, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.4235, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.3840, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.4656, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.3076, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.9686, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.9386, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.7965, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.5350, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.6344, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.0087, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.1317, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.7840, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.8785, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.3595, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.4893, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.0945, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.2419, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.6099, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(63.2155, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.9507, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.7103, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.7873, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.1087, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.8171, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.9084, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.6386, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.9134, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.9794, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.4580, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.8834, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4281567489114659, 'recall': 0.4351032448377581, 'f1': 0.43160204828090704, 'number': 678}, 'P': {'precision': 0.5475024485798237, 'recall': 0.6909765142150803, 'f1': 0.6109289617486339, 'number': 809}, 'overall_precision': 0.4994152046783626, 'overall_recall': 0.574310692669805, 'overall_f1': 0.53425086018142, 'overall_accuracy': 0.6923796681068883}
			------------EPOCH 27---------------
Loss:  tensor(27.2401, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.7832, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.9384, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.3601, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.1456, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.7507, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.5355, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.1217, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(59.1163, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.0617, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.3722, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.6360, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.2862, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.0925, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.4028, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.6425, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.6441, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.3900, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.9402, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(62.1867, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.9671, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.7573, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.4358, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.3278, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.2016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(74.8211, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.1567, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.4100, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.0701, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.5596, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.0796, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.6678, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.5499, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.3698, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.3597, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.8820, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.8519, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.41220238095238093, 'recall': 0.4085545722713864, 'f1': 0.4103703703703704, 'number': 678}, 'P': {'precision': 0.5386119257086999, 'recall': 0.681087762669963, 'f1': 0.601528384279476, 'number': 809}, 'overall_precision': 0.48849557522123893, 'overall_recall': 0.5568258238063215, 'overall_f1': 0.5204274041483343, 'overall_accuracy': 0.6943060792030162}
			------------EPOCH 28---------------
Loss:  tensor(26.0971, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.1839, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.4874, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.0194, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.8155, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.6955, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.3061, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.7390, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(59.1869, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.0060, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.4225, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.6893, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.5707, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.0941, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.1118, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.4891, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.9076, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.0100, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.5567, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.1505, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.9095, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.7009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.3797, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.3262, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.9970, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(68.5782, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.1357, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.0550, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.7243, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.6702, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.7207, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.6154, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.8334, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.9110, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.5413, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.0940, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.6651, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.41467065868263475, 'recall': 0.4085545722713864, 'f1': 0.41158989598811296, 'number': 678}, 'P': {'precision': 0.5374015748031497, 'recall': 0.6749072929542645, 'f1': 0.5983561643835618, 'number': 809}, 'overall_precision': 0.48871733966745845, 'overall_recall': 0.5534633490248824, 'overall_f1': 0.5190791548407443, 'overall_accuracy': 0.6950491234829512}
			------------EPOCH 29---------------
Loss:  tensor(22.8330, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.4237, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.1407, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.6548, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.6922, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.4786, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.2593, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.1180, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.6421, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.4525, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.6064, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.8512, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.5219, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.7515, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.2586, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.6554, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.8155, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.2988, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.7877, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.8211, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.2164, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.5245, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.7494, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.1118, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.9263, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.7901, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.7571, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.7764, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.6464, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.4013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.3189, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.6382, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.3882, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.3355, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.4323, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.8388, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.8684, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4171686746987952, 'recall': 0.4085545722713864, 'f1': 0.41281669150521605, 'number': 678}, 'P': {'precision': 0.5450941526263627, 'recall': 0.6798516687268232, 'f1': 0.6050605060506051, 'number': 809}, 'overall_precision': 0.4943215780035864, 'overall_recall': 0.5561533288500337, 'overall_f1': 0.5234177215189874, 'overall_accuracy': 0.6965352120428213}
			------------EPOCH 30---------------
Loss:  tensor(18.3133, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.9098, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.1607, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.7168, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.1627, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.1534, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.9391, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.1949, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(60.8452, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.5391, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.5821, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.4907, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.9040, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.1874, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.2750, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.8164, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.2091, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.7671, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.1380, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.2543, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.1998, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.6518, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.8574, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.7301, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.7106, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.8014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.8506, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.0641, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.8997, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.1156, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.4158, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.4835, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.1772, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.7196, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.7171, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.6303, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.8840, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.41220238095238093, 'recall': 0.4085545722713864, 'f1': 0.4103703703703704, 'number': 678}, 'P': {'precision': 0.5409674234945706, 'recall': 0.6773794808405439, 'f1': 0.601536772777168, 'number': 809}, 'overall_precision': 0.4896142433234421, 'overall_recall': 0.554808338937458, 'overall_f1': 0.5201765447667086, 'overall_accuracy': 0.6946363211052096}


		-------------RUN 5-----------
			------------EPOCH 1---------------
Loss:  tensor(2973.2166, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3117.0620, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1435.3547, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2948.1631, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2381.8091, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2173.9966, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1149.8484, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2254.4836, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3327.0303, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1720.6589, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(746.9901, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1859.8396, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2018.5801, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3018.3047, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2443.0654, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2272.9326, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(704.7090, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2176.7905, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1611.5431, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1252.6829, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(909.1063, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1621.8130, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1842.1544, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1326.9399, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(472.9734, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1042.2571, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2481.8252, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1962.2822, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2852.0620, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1480.9949, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1078.1835, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1943.0061, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2638.1660, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2817.7786, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2063.8989, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.09821428571428571, 'recall': 0.08138101109741061, 'f1': 0.08900876601483479, 'number': 811}, 'P': {'precision': 0.19972260748959778, 'recall': 0.16551724137931034, 'f1': 0.1810182275298554, 'number': 870}, 'overall_precision': 0.1507537688442211, 'overall_recall': 0.12492563950029745, 'overall_f1': 0.13662979830839297, 'overall_accuracy': 0.4829541393686718}
			------------EPOCH 2---------------
Loss:  tensor(2053.1038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2119.4478, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1009.8203, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2088.5898, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1745.5720, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1605.3943, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(830.1267, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1527.8383, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2599.2520, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1332.8369, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(569.2692, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1477.6382, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1665.0417, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2464.5181, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1962.9388, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1871.1622, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(516.8008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1829.6874, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1354.5786, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1037.8422, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(791.1854, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1345.4845, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1634.0042, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1154.7212, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(359.8855, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(967.2671, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2314.0293, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1800.2704, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2761.6528, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1509.4824, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1087.2905, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1970.9885, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2631.7278, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2522.8721, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1685.1089, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.14888888888888888, 'recall': 0.08261405672009864, 'f1': 0.10626486915146707, 'number': 811}, 'P': {'precision': 0.2215568862275449, 'recall': 0.29770114942528736, 'f1': 0.25404610102991665, 'number': 870}, 'overall_precision': 0.20135886349598517, 'overall_recall': 0.19393218322427128, 'overall_f1': 0.19757575757575757, 'overall_accuracy': 0.5273377010125074}
			------------EPOCH 3---------------
Loss:  tensor(1652.1482, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1664.4989, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(803.9016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1672.3361, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1522.6545, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1389.9893, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(747.9002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1386.1826, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2392.8960, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1176.0820, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(461.7322, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1320.6821, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1502.5486, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2183.3120, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1927.7480, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1802.3037, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(414.4348, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1670.1846, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1149.5863, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(864.7974, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(635.5328, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1154.9005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1451.6837, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(996.5231, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(321.7968, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(905.2457, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1889.2708, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1614.5419, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2375.8525, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1336.2881, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(990.6458, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1762.4570, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2170.4578, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2307.9265, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1405.8066, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.15571776155717762, 'recall': 0.07891491985203453, 'f1': 0.10474631751227496, 'number': 811}, 'P': {'precision': 0.2539912917271408, 'recall': 0.40229885057471265, 'f1': 0.3113879003558719, 'number': 870}, 'overall_precision': 0.23141419787590833, 'overall_recall': 0.2462819750148721, 'overall_f1': 0.23861671469740633, 'overall_accuracy': 0.5619773674806432}
			------------EPOCH 4---------------
Loss:  tensor(1499.8820, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1250.4387, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(647.7988, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1361.4767, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1279.1206, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1177.8779, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(587.7241, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1063.4573, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2031.6646, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1031.4380, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(405.5174, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1121.3755, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1373.9154, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1858.3472, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1719.3232, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1541.4156, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(345.8733, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1394.6409, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(971.8832, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(782.6858, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(546.4197, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(976.9476, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1267.8431, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(861.9520, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(284.0473, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(793.4640, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1339.3345, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1445.5432, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1805.3660, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(919.9706, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(783.5320, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1309.0376, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1784.9351, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1961.5370, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1159.3842, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.21407624633431085, 'recall': 0.09001233045622688, 'f1': 0.1267361111111111, 'number': 811}, 'P': {'precision': 0.2523961661341853, 'recall': 0.4540229885057471, 'f1': 0.32443531827515404, 'number': 870}, 'overall_precision': 0.24554039874081846, 'overall_recall': 0.27840571088637717, 'overall_f1': 0.2609422916085866, 'overall_accuracy': 0.5787730792138177}
			------------EPOCH 5---------------
Loss:  tensor(1228.6199, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(958.7224, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(555.5430, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1167.5820, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1122.5416, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1046.5426, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(466.7381, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(851.5293, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1646.4199, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(932.3428, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(319.0249, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(877.4388, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1120.0872, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1499.9902, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1371.1274, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1182.3810, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(288.5045, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1148.1324, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(790.0668, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(653.7518, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(417.3345, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(822.7516, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1049.7017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(728.6871, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(238.8987, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(643.7087, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(914.9780, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1241.6383, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1309.0232, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(513.7802, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(541.4155, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(829.6864, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1332.9319, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1561.7014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(953.8722, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3333333333333333, 'recall': 0.15166461159062886, 'f1': 0.20847457627118643, 'number': 811}, 'P': {'precision': 0.2675693101225016, 'recall': 0.47701149425287354, 'f1': 0.34283353985956216, 'number': 870}, 'overall_precision': 0.28020833333333334, 'overall_recall': 0.32004759071980965, 'overall_f1': 0.2988058872535407, 'overall_accuracy': 0.6047647409172127}
			------------EPOCH 6---------------
Loss:  tensor(948.2438, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(675.2739, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(431.9120, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(922.7916, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(974.0561, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(928.5453, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(369.3766, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(650.0974, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1218.1125, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(771.8492, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(277.2838, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(685.1098, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(865.8174, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1063.4601, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1037.0414, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(812.7106, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(209.9976, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(864.3347, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(585.2896, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(493.9395, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(304.7458, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(661.5988, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(800.4330, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(598.0552, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(202.2280, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(507.9025, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(527.6495, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(983.6567, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(907.2585, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(324.1938, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(390.8269, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(463.0718, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(997.7212, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1218.4874, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(767.4026, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.43073047858942065, 'recall': 0.21085080147965474, 'f1': 0.2831125827814569, 'number': 811}, 'P': {'precision': 0.36711409395973155, 'recall': 0.628735632183908, 'f1': 0.46355932203389827, 'number': 870}, 'overall_precision': 0.3804981452040276, 'overall_recall': 0.4271267102914932, 'overall_f1': 0.4024663677130045, 'overall_accuracy': 0.6501250744490769}
			------------EPOCH 7---------------
Loss:  tensor(735.6493, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(463.1371, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(322.2322, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(686.8665, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(814.7512, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(724.3457, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(304.1528, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(492.2504, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(884.8867, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(624.3259, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(230.4434, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(539.5645, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(631.1741, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(741.0062, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(751.5638, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(567.1672, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(151.6312, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(659.1426, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(411.1246, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(363.2547, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(264.4774, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(576.9532, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(641.2136, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(508.4572, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(164.8529, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(385.3001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(349.1174, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(774.1699, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(533.6661, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(240.2788, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(231.2533, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(265.6010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(667.0154, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(943.4312, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(575.4703, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.46745562130177515, 'recall': 0.19482120838471023, 'f1': 0.2750217580504787, 'number': 811}, 'P': {'precision': 0.3932820153953814, 'recall': 0.6459770114942529, 'f1': 0.4889082209656373, 'number': 870}, 'overall_precision': 0.4074702886247878, 'overall_recall': 0.4283164782867341, 'overall_f1': 0.41763341067285387, 'overall_accuracy': 0.6532698034544372}
			------------EPOCH 8---------------
Loss:  tensor(520.6451, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(273.9458, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(241.4442, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(489.4223, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(709.7303, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(564.2983, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(246.4405, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(352.2082, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(720.1323, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(506.9549, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(210.4448, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(529.3388, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(455.0659, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(569.5789, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(569.5988, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(442.6086, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(95.9919, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(423.5359, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(313.8131, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(241.8475, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(188.5777, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(452.9287, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(457.7063, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(388.4818, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(166.0712, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(462.9873, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(302.0553, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(820.0610, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(659.9028, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(227.9747, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(256.9056, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(302.5999, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(654.7832, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(907.2400, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(622.9753, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4373795761078998, 'recall': 0.279901356350185, 'f1': 0.34135338345864663, 'number': 811}, 'P': {'precision': 0.39032697547683926, 'recall': 0.6586206896551724, 'f1': 0.4901625320786997, 'number': 870}, 'overall_precision': 0.40261701056869653, 'overall_recall': 0.4759071980963712, 'overall_f1': 0.4362050163576881, 'overall_accuracy': 0.6504824300178678}
			------------EPOCH 9---------------
Loss:  tensor(373.6092, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(203.3687, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(189.1974, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(356.2150, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(572.7584, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(452.2304, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(200.2506, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(271.7739, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(848.1476, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(546.3053, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(278.7215, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(527.0469, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(738.3406, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1058.0859, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(800.6028, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(814.7805, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(221.1338, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(944.2401, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(623.7354, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(709.5012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(200.6663, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(662.4318, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(653.1152, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(638.3823, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(82.4749, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(325.1252, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(247.6117, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(585.7274, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(423.7769, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(186.5600, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(207.3174, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(235.7482, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(995.8825, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1191.4644, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1201.7528, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.27009294696555497, 'recall': 0.6091245376078915, 'f1': 0.37424242424242427, 'number': 811}, 'P': {'precision': 0.6851851851851852, 'recall': 0.12758620689655173, 'f1': 0.21511627906976746, 'number': 870}, 'overall_precision': 0.30386740331491713, 'overall_recall': 0.3599048185603807, 'overall_f1': 0.32952069716775595, 'overall_accuracy': 0.5335318642048839}
			------------EPOCH 10---------------
Loss:  tensor(1518.0974, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1280.8354, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(462.6450, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1092.0034, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(917.8692, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(891.8107, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(481.0941, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(973.2284, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1257.8811, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1024.8383, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(304.9125, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(563.5983, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(828.5007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1005.5642, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(716.1283, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1223.6887, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(178.5726, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(816.0906, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(609.0184, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(428.7212, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(219.9832, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(646.9147, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(727.9316, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(776.7134, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(126.1223, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(606.6309, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(791.3146, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1222.1689, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1154.1721, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(366.5711, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(505.3777, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(605.3947, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(735.7921, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1678.0658, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(925.8671, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.2536764705882353, 'recall': 0.08508014796547472, 'f1': 0.12742382271468145, 'number': 811}, 'P': {'precision': 0.3942486085343228, 'recall': 0.4885057471264368, 'f1': 0.43634496919917864, 'number': 870}, 'overall_precision': 0.36592592592592593, 'overall_recall': 0.29387269482450923, 'overall_f1': 0.32596502804355, 'overall_accuracy': 0.5668374032162001}
			------------EPOCH 11---------------
Loss:  tensor(859.1488, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(600.4653, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(480.0949, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(947.4435, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(789.7118, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(807.5245, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(384.5626, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(589.8942, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(831.6827, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(533.1437, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(206.1540, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(464.9324, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(523.1740, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(600.7399, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(618.5179, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(459.1735, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(134.2865, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(519.6835, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(378.0337, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(253.2849, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(185.5063, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(454.4202, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(489.6346, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(416.8939, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(156.5634, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(374.5275, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(578.9827, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(736.2399, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(780.4895, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(383.3249, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(309.5088, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(559.7996, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1338.4128, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1142.3483, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(650.2840, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3296032553407935, 'recall': 0.3995067817509248, 'f1': 0.36120401337792646, 'number': 811}, 'P': {'precision': 0.40890008396305627, 'recall': 0.5597701149425287, 'f1': 0.472586123241145, 'number': 870}, 'overall_precision': 0.3730450781968721, 'overall_recall': 0.4824509220701963, 'overall_f1': 0.42075226977950714, 'overall_accuracy': 0.6165098272781417}
			------------EPOCH 12---------------
Loss:  tensor(518.1476, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(361.2001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(207.7544, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(455.9719, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(522.0493, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(465.5019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(236.8646, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(408.7432, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(694.8897, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(408.7814, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(188.3992, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(342.6561, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(542.4490, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(555.4764, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(603.3828, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(320.5607, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(118.5839, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(396.0729, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(279.8241, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(210.5846, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(141.7020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(377.3453, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(349.6069, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(290.1556, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(78.6154, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(304.5703, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(176.3049, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(402.6378, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(368.0312, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(150.1053, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(251.9764, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(169.4657, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(357.6331, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(719.1828, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(367.9310, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5286624203821656, 'recall': 0.3070283600493218, 'f1': 0.38845553822152884, 'number': 811}, 'P': {'precision': 0.43914081145584727, 'recall': 0.6344827586206897, 'f1': 0.5190409026798307, 'number': 870}, 'overall_precision': 0.4635416666666667, 'overall_recall': 0.4765020820939917, 'overall_f1': 0.4699325315341743, 'overall_accuracy': 0.6884335914234664}
			------------EPOCH 13---------------
Loss:  tensor(243.6487, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(141.2005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(131.6632, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(316.0185, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(457.9047, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(289.0811, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(106.2531, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(272.9171, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(341.8910, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(210.3479, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(118.9730, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(178.0628, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(271.5069, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(293.6199, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(342.2958, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(224.1700, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.6349, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(226.5483, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(225.1698, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(103.0286, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(100.8800, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(260.2103, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(190.8894, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(206.4712, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.8345, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(193.4283, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(77.0192, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(208.2378, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(162.7737, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(104.6027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(112.9296, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(94.5918, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(212.5439, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(347.2386, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(231.3068, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4303030303030303, 'recall': 0.43773119605425403, 'f1': 0.4339853300733496, 'number': 811}, 'P': {'precision': 0.4503311258278146, 'recall': 0.5471264367816092, 'f1': 0.4940321743642969, 'number': 870}, 'overall_precision': 0.44155154091392135, 'overall_recall': 0.4943486020226056, 'overall_f1': 0.46646084760033674, 'overall_accuracy': 0.6829303156640858}
			------------EPOCH 14---------------
Loss:  tensor(136.0013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(71.3563, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(77.4270, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(182.7769, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(325.4158, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(226.6391, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(87.0959, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(194.8643, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(264.6975, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(168.6323, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(101.7680, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(130.2036, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(194.9075, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(228.7902, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(287.7264, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(145.9682, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.2171, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(144.4515, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(158.3023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(76.0719, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(77.1797, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(196.8532, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(138.8654, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(145.8687, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.1353, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(135.5605, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.2610, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(125.9553, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(118.9520, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(89.6910, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(93.9158, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(76.2812, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(138.6780, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(225.8705, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(151.6458, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5165016501650165, 'recall': 0.38594327990135635, 'f1': 0.4417784050811574, 'number': 811}, 'P': {'precision': 0.45491803278688525, 'recall': 0.6379310344827587, 'f1': 0.5311004784688996, 'number': 870}, 'overall_precision': 0.47535596933187296, 'overall_recall': 0.5163593099345628, 'overall_f1': 0.4950099800399202, 'overall_accuracy': 0.6828588445503275}
			------------EPOCH 15---------------
Loss:  tensor(107.4091, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.4779, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.3750, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(145.3559, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(242.8692, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(156.1033, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.8849, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(122.1287, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(150.3099, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(129.9330, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(67.0521, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(78.3068, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(119.2097, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(147.9590, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(203.7045, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(104.2338, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.9135, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(93.0005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(122.6753, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.1062, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.3744, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(133.2311, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(99.9392, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(126.6302, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.4248, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(104.0670, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.1706, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(84.7362, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(85.2995, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(71.4071, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(65.1784, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.8929, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(86.6425, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(148.0817, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(96.6774, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4837837837837838, 'recall': 0.44143033292231815, 'f1': 0.4616376531270148, 'number': 811}, 'P': {'precision': 0.44454887218045114, 'recall': 0.5436781609195402, 'f1': 0.48914167528438474, 'number': 870}, 'overall_precision': 0.4606430155210643, 'overall_recall': 0.4943486020226056, 'overall_f1': 0.4769010043041607, 'overall_accuracy': 0.67776057176891}
			------------EPOCH 16---------------
Loss:  tensor(79.5039, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.9391, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.5014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(97.6508, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(185.6101, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(104.1508, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.4854, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(86.4940, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(114.2321, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(88.7678, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.6224, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.2910, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(76.5002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(117.6189, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(165.8214, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(76.4409, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.8457, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(62.8371, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(90.0577, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.1289, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.3035, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(107.0060, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(71.8969, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(86.7685, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.7247, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(70.6392, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.8544, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(59.3736, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(69.3034, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.7703, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.7765, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.0695, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(65.6185, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(106.2713, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(64.7923, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.47280334728033474, 'recall': 0.41800246609124536, 'f1': 0.443717277486911, 'number': 811}, 'P': {'precision': 0.4339788732394366, 'recall': 0.5666666666666667, 'f1': 0.4915254237288135, 'number': 870}, 'overall_precision': 0.4490016189962223, 'overall_recall': 0.49494348602022603, 'overall_f1': 0.47085455574419915, 'overall_accuracy': 0.6748302561048243}
			------------EPOCH 17---------------
Loss:  tensor(57.6683, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.6702, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.4581, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(72.7988, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(139.6612, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(73.5520, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.0472, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(68.0866, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(86.6982, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(62.0929, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.7941, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.2156, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.4785, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(98.2571, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(137.9665, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.6551, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.2882, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.5742, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(73.0784, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.7850, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.8003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(80.7623, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.1382, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(65.2882, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.1676, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.0567, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.9848, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.4896, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(61.6270, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.0626, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.2670, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.0750, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.0322, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(83.3396, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.4002, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.469187675070028, 'recall': 0.4130702836004932, 'f1': 0.43934426229508194, 'number': 811}, 'P': {'precision': 0.4288150042625746, 'recall': 0.5781609195402299, 'f1': 0.49241311796377873, 'number': 870}, 'overall_precision': 0.4440911499735029, 'overall_recall': 0.49851279000594884, 'overall_f1': 0.4697309417040359, 'overall_accuracy': 0.6741155449672424}
			------------EPOCH 18---------------
Loss:  tensor(40.4121, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.9751, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.4897, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(58.5492, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(115.9732, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.4022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.0721, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(56.8909, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(71.1006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.5045, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.0405, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.0067, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.3675, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(83.9077, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(119.6945, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.8305, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.1150, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.9115, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.0159, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.5092, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.7754, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.9204, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.5130, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(50.9012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.7422, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.9489, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.4666, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.5681, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(55.1933, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.8653, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.7310, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.0046, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.3952, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(66.8839, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.0937, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.45751633986928103, 'recall': 0.4315659679408138, 'f1': 0.4441624365482233, 'number': 811}, 'P': {'precision': 0.4369900271985494, 'recall': 0.5540229885057472, 'f1': 0.48859604662949824, 'number': 870}, 'overall_precision': 0.44539614561027835, 'overall_recall': 0.49494348602022603, 'overall_f1': 0.4688644688644688, 'overall_accuracy': 0.6730434782608695}
			------------EPOCH 19---------------
Loss:  tensor(30.7183, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.9064, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.0818, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.5006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(101.3627, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.7644, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.1134, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.4618, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(61.0162, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.4315, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.7684, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.7572, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(33.7076, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(71.7170, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(102.1196, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.1515, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.3069, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.5545, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.1854, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.0791, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.7403, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.7658, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.1620, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.2549, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.9295, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.4176, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.3975, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.1617, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(51.2425, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.1720, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.0977, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.6160, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.7613, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(57.7678, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.0403, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4608108108108108, 'recall': 0.4204685573366215, 'f1': 0.43971631205673767, 'number': 811}, 'P': {'precision': 0.4346689895470383, 'recall': 0.5735632183908046, 'f1': 0.49454905847373637, 'number': 870}, 'overall_precision': 0.4449152542372881, 'overall_recall': 0.4997025580011898, 'overall_f1': 0.47072008966096945, 'overall_accuracy': 0.6729481834425253}
			------------EPOCH 20---------------
Loss:  tensor(26.0578, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.5000, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.3662, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.7177, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(88.1091, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.1643, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.7191, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.3003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(53.2150, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.3403, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.0219, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.2337, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.2683, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(61.6945, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(88.7785, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.5698, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.7630, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.8138, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.6803, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.8266, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.3248, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.4537, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.7828, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.6951, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.3767, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.0675, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.6079, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.3184, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.6835, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.3515, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.5513, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.7608, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.6702, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(49.2910, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.2522, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.45442536327608984, 'recall': 0.4241676942046856, 'f1': 0.4387755102040816, 'number': 811}, 'P': {'precision': 0.45135135135135135, 'recall': 0.5758620689655173, 'f1': 0.5060606060606061, 'number': 870}, 'overall_precision': 0.45259775040171396, 'overall_recall': 0.5026769779892921, 'overall_f1': 0.4763246899661781, 'overall_accuracy': 0.6739249553305539}
			------------EPOCH 21---------------
Loss:  tensor(22.5588, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.4435, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.6907, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.8657, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(80.0368, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.6337, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.9730, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.0966, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.4512, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.4448, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.4973, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.5771, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.2136, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.7500, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(76.4670, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.7414, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.0558, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.9155, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.0400, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.0861, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.8971, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.7829, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.6620, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.0314, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.1908, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.6273, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.3818, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.3390, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.0997, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.0557, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.6344, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.5819, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.2209, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.6770, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.7967, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4445910290237467, 'recall': 0.4155363748458693, 'f1': 0.42957297641810066, 'number': 811}, 'P': {'precision': 0.4265975820379965, 'recall': 0.5678160919540229, 'f1': 0.4871794871794871, 'number': 870}, 'overall_precision': 0.4337160751565762, 'overall_recall': 0.4943486020226056, 'overall_f1': 0.46205170975813176, 'overall_accuracy': 0.672852888624181}
			------------EPOCH 22---------------
Loss:  tensor(20.7579, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.0160, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.2674, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.2755, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(72.3208, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.0550, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.8320, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.6434, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.3300, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.4679, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.1711, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.9426, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.7161, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(47.2328, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(69.1643, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.5070, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.3561, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.5389, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.7708, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.8726, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.5480, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.5665, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.7915, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.5958, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.0856, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.9919, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.4141, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.4404, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(46.0402, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.0219, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.1016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.6286, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.5533, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.9743, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.9471, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.44357976653696496, 'recall': 0.4217016029593095, 'f1': 0.43236409608091025, 'number': 811}, 'P': {'precision': 0.4413854351687389, 'recall': 0.5712643678160919, 'f1': 0.497995991983968, 'number': 870}, 'overall_precision': 0.4422772799156563, 'overall_recall': 0.4991076740035693, 'overall_f1': 0.4689770821688094, 'overall_accuracy': 0.6753782013103038}
			------------EPOCH 23---------------
Loss:  tensor(19.2670, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.5045, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.4940, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.6750, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(66.5613, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.4392, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.4488, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.7995, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.7027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.4443, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.7929, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.1381, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.4269, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(39.7418, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(59.9159, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.0474, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.1855, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.3794, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.9273, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.4174, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.8995, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.7656, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.6952, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.3787, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.0850, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.5290, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.4387, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.5914, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.2591, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.8361, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.9590, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.8875, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.9816, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.7991, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.9237, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.43617021276595747, 'recall': 0.40443896424167697, 'f1': 0.4197056941778631, 'number': 811}, 'P': {'precision': 0.4314236111111111, 'recall': 0.5712643678160919, 'f1': 0.4915924826904055, 'number': 870}, 'overall_precision': 0.4332983193277311, 'overall_recall': 0.4907792980368828, 'overall_f1': 0.46025104602510464, 'overall_accuracy': 0.669398451459202}
			------------EPOCH 24---------------
Loss:  tensor(18.2816, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.5734, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.5120, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.3457, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(59.9061, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.9839, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.5793, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.1516, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(37.4353, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.0633, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.9107, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.3721, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.1778, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(36.0039, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(52.9169, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.3862, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.2353, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.2691, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.8280, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.5746, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.1968, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.2319, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.3122, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.0714, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.4589, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.2072, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.2783, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.1193, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(44.7050, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.0896, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.8706, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.1327, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.6738, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.1412, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.2733, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4384105960264901, 'recall': 0.4081381011097411, 'f1': 0.42273307790549175, 'number': 811}, 'P': {'precision': 0.44014084507042256, 'recall': 0.5747126436781609, 'f1': 0.4985044865403789, 'number': 870}, 'overall_precision': 0.4394500264410365, 'overall_recall': 0.4943486020226056, 'overall_f1': 0.46528555431131013, 'overall_accuracy': 0.6692078618225134}
			------------EPOCH 25---------------
Loss:  tensor(17.6596, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.2948, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.6854, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(54.5901, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.5084, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.7737, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.8569, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.8121, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.6988, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.5989, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.4532, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.3252, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.9486, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(45.7913, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.4936, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.6279, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.0192, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.9067, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.5347, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.8525, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.8918, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.7969, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.8835, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.8296, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.4237, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.3359, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.7351, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.8838, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.4931, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.9367, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.6223, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.6480, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(27.1669, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.7746, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4410187667560322, 'recall': 0.40567200986436497, 'f1': 0.4226075786769429, 'number': 811}, 'P': {'precision': 0.4360313315926893, 'recall': 0.5758620689655173, 'f1': 0.4962852897473997, 'number': 870}, 'overall_precision': 0.43799472295514513, 'overall_recall': 0.49375371802498513, 'overall_f1': 0.46420581655480986, 'overall_accuracy': 0.6713758189398451}
			------------EPOCH 26---------------
Loss:  tensor(16.5002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.4686, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.6809, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.3677, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(48.7721, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.8287, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.9211, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.4705, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(32.4308, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.3376, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.5792, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.9798, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.9270, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.5436, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.6328, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.9423, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(4.0155, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.4907, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.6068, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.7353, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.6255, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.7822, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.4009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.0397, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.0490, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.6484, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.6688, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.3127, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.2140, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.2423, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.1253, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.1379, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.0848, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.9738, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.7243, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.45, 'recall': 0.41060419235511714, 'f1': 0.4294003868471954, 'number': 811}, 'P': {'precision': 0.4386422976501306, 'recall': 0.5793103448275863, 'f1': 0.49925705794948, 'number': 870}, 'overall_precision': 0.44309158284806777, 'overall_recall': 0.49791790600832836, 'overall_f1': 0.4689075630252101, 'overall_accuracy': 0.671304347826087}
			------------EPOCH 27---------------
Loss:  tensor(15.8630, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.7900, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.7165, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.1427, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(43.6698, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.8432, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.3047, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.7323, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.2255, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.5903, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.8460, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.2951, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.1117, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(25.7740, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(35.5789, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.6577, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.5776, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.5386, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.9539, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.9346, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.2840, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.8121, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.0328, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.6667, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.5313, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.0557, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.0016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.2335, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(42.6011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.0711, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.6739, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.7451, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.7380, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.2625, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.9703, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.45108695652173914, 'recall': 0.4093711467324291, 'f1': 0.42921784098254684, 'number': 811}, 'P': {'precision': 0.4358752166377816, 'recall': 0.5781609195402299, 'f1': 0.49703557312252966, 'number': 870}, 'overall_precision': 0.4417989417989418, 'overall_recall': 0.49672813801308746, 'overall_f1': 0.46765611873424806, 'overall_accuracy': 0.6709708159618821}
			------------EPOCH 28---------------
Loss:  tensor(15.1048, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.1755, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.9307, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.7483, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(38.6104, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.7274, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.6189, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.1332, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.0097, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.7968, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.2999, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.3960, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.0483, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.6942, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(31.9244, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.5453, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.1897, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(24.0037, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.8687, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.2667, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.2837, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.0047, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.0671, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.5600, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.0370, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.5744, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.3191, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.1876, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.9150, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.9356, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.8636, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.4320, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.6753, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.8051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.2338, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.45121951219512196, 'recall': 0.41060419235511714, 'f1': 0.42995480955455134, 'number': 811}, 'P': {'precision': 0.4358752166377816, 'recall': 0.5781609195402299, 'f1': 0.49703557312252966, 'number': 870}, 'overall_precision': 0.4418604651162791, 'overall_recall': 0.49732302201070794, 'overall_f1': 0.4679541001959138, 'overall_accuracy': 0.6710899344848124}
			------------EPOCH 29---------------
Loss:  tensor(14.5855, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.6628, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.2792, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.3862, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(34.2408, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.0144, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.0897, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.6189, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(29.0854, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.1415, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.9312, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.7386, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.1653, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(21.9440, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.9544, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.5391, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.8861, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(23.3026, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.8347, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.6501, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.4277, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.3708, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.1460, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.6244, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.5955, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(16.3184, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.6640, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.2217, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(41.2095, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.8782, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.4607, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.1603, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.7987, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.6393, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.6388, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4506089309878214, 'recall': 0.41060419235511714, 'f1': 0.4296774193548387, 'number': 811}, 'P': {'precision': 0.43301642178046673, 'recall': 0.5758620689655173, 'f1': 0.4943265910212136, 'number': 870}, 'overall_precision': 0.439873417721519, 'overall_recall': 0.496133254015467, 'overall_f1': 0.4663125524182276, 'overall_accuracy': 0.6698511018463371}
			------------EPOCH 30---------------
Loss:  tensor(14.0718, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.1887, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.6942, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.3276, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(30.3436, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.2227, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.5570, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.2360, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(28.1854, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.5737, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.7408, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.0723, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.3114, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(20.4842, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(26.8225, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(12.6867, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5966, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(22.7238, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(10.9906, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.0861, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.7002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.4854, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(18.3791, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.8226, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.2016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.2730, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(11.0650, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(17.2587, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(40.5483, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(14.9521, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(5.1888, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.8701, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(13.0949, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(19.7023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(15.1211, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.45135135135135135, 'recall': 0.4118372379778052, 'f1': 0.4306898774983881, 'number': 811}, 'P': {'precision': 0.439066551426102, 'recall': 0.5839080459770115, 'f1': 0.501233349777997, 'number': 870}, 'overall_precision': 0.44385872430152873, 'overall_recall': 0.5008923259964307, 'overall_f1': 0.4706539966461711, 'overall_accuracy': 0.6699225729600953}
