Tokenizer: bert-base-cased Model: bert-base-cased
	Train size: 80 Test size: 20


		-------------RUN 1-----------
			------------EPOCH 1---------------
Loss:  tensor(3.3341, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.1149, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4798, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1532, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5915, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2654, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5577, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8708, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.7977, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.4947, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.4615, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.1879, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2587, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.4504, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0150, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.3542, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2401, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4589, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2330, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3219, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4636, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.0432, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.9736, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.9422, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.0921, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.9241, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2458, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.9068, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.7020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0943, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1398, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.7220, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6999, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4493, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.8237, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1396, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.7499, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0860, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.7396, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1955, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.2904191616766467, 'recall': 0.20250521920668058, 'f1': 0.23862238622386225, 'number': 479}, 'P': {'precision': 0.42917726887192537, 'recall': 0.593200468933177, 'f1': 0.4980314960629921, 'number': 853}, 'overall_precision': 0.3985459352280238, 'overall_recall': 0.4527027027027027, 'overall_f1': 0.42390158172231984, 'overall_accuracy': 0.7399234541322808}
			------------EPOCH 2---------------
Loss:  tensor(1.5802, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4637, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2647, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6427, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5448, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3159, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3960, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4776, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0607, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4730, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3405, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5983, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4354, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5515, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6161, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3783, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4184, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9144, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4787, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2000, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3063, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2566, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3090, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2346, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3059, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1852, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1474, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1652, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9763, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8229, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0913, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2230, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1496, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3184, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2634, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7391, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2676, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0607, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2159, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8583, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.38286713286713286, 'recall': 0.4572025052192067, 'f1': 0.41674595623215993, 'number': 479}, 'P': {'precision': 0.5313152400835073, 'recall': 0.5967174677608441, 'f1': 0.5621203754831585, 'number': 853}, 'overall_precision': 0.47581699346405226, 'overall_recall': 0.5465465465465466, 'overall_f1': 0.5087351502445842, 'overall_accuracy': 0.8054359526372443}
			------------EPOCH 3---------------
Loss:  tensor(1.0426, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0406, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1811, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4433, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1044, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8987, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9428, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3755, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7919, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9490, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2729, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0233, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0477, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3949, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9000, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9511, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5872, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9684, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1402, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2231, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8284, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8762, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8235, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8434, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8247, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0868, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7510, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6622, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6080, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0683, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0232, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9178, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2512, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8765, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5118, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9558, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0467, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8597, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5720, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.38461538461538464, 'recall': 0.4592901878914405, 'f1': 0.41864890580399616, 'number': 479}, 'P': {'precision': 0.5887179487179487, 'recall': 0.6729191090269636, 'f1': 0.6280087527352297, 'number': 853}, 'overall_precision': 0.513251454427925, 'overall_recall': 0.5960960960960962, 'overall_f1': 0.5515804098645363, 'overall_accuracy': 0.8225092692261691}
			------------EPOCH 4---------------
Loss:  tensor(0.7394, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7928, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1411, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3478, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8585, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6907, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6681, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2641, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6092, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8617, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7296, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1053, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8764, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8332, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2812, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6898, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6453, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3794, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6498, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0886, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1897, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5982, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6835, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6580, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5249, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6250, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0511, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5234, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4385, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4292, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0459, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8994, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7889, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2366, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6764, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4741, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8178, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0388, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5945, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3903, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.39267015706806285, 'recall': 0.4697286012526096, 'f1': 0.42775665399239543, 'number': 479}, 'P': {'precision': 0.5018975332068312, 'recall': 0.6201641266119577, 'f1': 0.5547981122181437, 'number': 853}, 'overall_precision': 0.4634296250768285, 'overall_recall': 0.566066066066066, 'overall_f1': 0.5096316323082123, 'overall_accuracy': 0.8179643583303433}
			------------EPOCH 5---------------
Loss:  tensor(0.5470, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6040, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1283, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2483, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9916, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5761, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5039, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1939, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5502, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6442, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4458, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8620, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7428, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5849, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1926, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6467, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5902, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3948, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5672, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0775, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1560, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5150, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4787, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4540, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3690, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3964, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0455, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4154, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4793, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3615, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0223, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8516, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6890, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1929, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4746, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3320, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6273, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0367, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3938, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2933, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4444444444444444, 'recall': 0.4592901878914405, 'f1': 0.4517453798767967, 'number': 479}, 'P': {'precision': 0.5740384615384615, 'recall': 0.6998827667057445, 'f1': 0.6307448494453249, 'number': 853}, 'overall_precision': 0.5322475570032573, 'overall_recall': 0.6133633633633634, 'overall_f1': 0.5699337286362051, 'overall_accuracy': 0.8224494677670136}
			------------EPOCH 6---------------
Loss:  tensor(0.4547, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4118, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0676, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1546, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5361, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3640, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3111, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1432, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4153, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4429, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2584, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7706, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5807, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3953, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1770, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3423, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3988, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2110, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2884, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0481, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1273, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2975, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4488, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3652, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2749, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3585, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0230, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2305, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2099, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2169, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0130, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6581, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6657, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1851, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3603, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2264, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5069, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0164, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2937, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2201, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3445287107258938, 'recall': 0.6638830897703549, 'f1': 0.45363766048502135, 'number': 479}, 'P': {'precision': 0.6339144215530903, 'recall': 0.46893317702227433, 'f1': 0.5390835579514824, 'number': 853}, 'overall_precision': 0.462033462033462, 'overall_recall': 0.539039039039039, 'overall_f1': 0.49757449757449757, 'overall_accuracy': 0.7590599210620739}
			------------EPOCH 7---------------
Loss:  tensor(0.5235, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5570, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0770, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1144, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5230, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2514, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2495, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0869, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4461, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4094, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7896, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5522, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3920, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1451, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2615, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1070, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1761, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0221, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1151, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2198, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2711, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2218, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2057, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1941, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0173, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2223, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2549, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2035, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0170, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8532, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5950, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1852, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1635, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1916, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4573, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0081, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6213, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4434, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.519650655021834, 'recall': 0.4968684759916493, 'f1': 0.5080042689434365, 'number': 479}, 'P': {'precision': 0.603112840466926, 'recall': 0.7268464243845252, 'f1': 0.659223817118554, 'number': 853}, 'overall_precision': 0.5773889636608345, 'overall_recall': 0.6441441441441441, 'overall_f1': 0.6089425124201561, 'overall_accuracy': 0.8224494677670136}
			------------EPOCH 8---------------
Loss:  tensor(0.1557, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1339, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0514, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0739, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4587, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1844, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2410, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1449, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3865, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4171, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2946, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6780, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5081, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3520, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1170, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2264, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2125, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1342, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2933, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0322, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1622, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2781, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3459, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2593, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1586, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2167, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0115, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1563, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1645, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1403, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0137, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7106, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5954, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1947, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2525, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2488, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4299, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0288, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1565, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1450, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5400943396226415, 'recall': 0.4780793319415449, 'f1': 0.5071982281284606, 'number': 479}, 'P': {'precision': 0.6085232903865213, 'recall': 0.7198124267291911, 'f1': 0.6595059076262083, 'number': 853}, 'overall_precision': 0.5882763433356595, 'overall_recall': 0.6328828828828829, 'overall_f1': 0.6097649186256782, 'overall_accuracy': 0.8182334648965435}
			------------EPOCH 9---------------
Loss:  tensor(0.1219, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1083, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0573, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0685, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5997, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1754, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1741, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0765, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4330, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2450, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1147, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5894, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4391, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2272, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1077, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1955, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1329, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0832, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1558, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0118, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0955, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1639, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2126, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1565, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1047, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1050, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0078, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0856, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0090, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5816, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4684, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1444, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0847, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0882, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3170, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0030, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0938, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.45841784989858014, 'recall': 0.4718162839248434, 'f1': 0.4650205761316873, 'number': 479}, 'P': {'precision': 0.6340725806451613, 'recall': 0.7373974208675264, 'f1': 0.6818428184281842, 'number': 853}, 'overall_precision': 0.5757575757575758, 'overall_recall': 0.6418918918918919, 'overall_f1': 0.6070287539936102, 'overall_accuracy': 0.8225092692261691}
			------------EPOCH 10---------------
Loss:  tensor(0.0661, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0473, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0278, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0523, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3538, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0978, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0820, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0453, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2759, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1805, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0665, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4912, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3353, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1805, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0766, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0626, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0567, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0385, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0622, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0042, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0687, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0971, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1653, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0666, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0645, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0790, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0036, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0610, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0670, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0924, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0077, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4300, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3755, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1044, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0504, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1359, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2367, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0492, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0567, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.49693251533742333, 'recall': 0.5073068893528184, 'f1': 0.5020661157024793, 'number': 479}, 'P': {'precision': 0.6375, 'recall': 0.7174677608440797, 'f1': 0.6751241036955322, 'number': 853}, 'overall_precision': 0.5900621118012422, 'overall_recall': 0.6418918918918919, 'overall_f1': 0.6148867313915857, 'overall_accuracy': 0.8242136108121039}
			------------EPOCH 11---------------
Loss:  tensor(0.0266, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0404, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0067, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0354, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2993, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0506, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0449, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0327, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2364, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1637, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0546, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4409, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2821, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1968, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0611, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0421, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0414, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0292, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0468, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0033, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0675, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0804, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1363, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0576, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0348, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0590, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0030, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0365, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0462, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0648, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0040, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4170, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3547, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0905, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0542, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1045, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1782, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0498, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0502, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.49473684210526314, 'recall': 0.4906054279749478, 'f1': 0.49266247379454925, 'number': 479}, 'P': {'precision': 0.6367567567567568, 'recall': 0.690504103165299, 'f1': 0.6625421822272217, 'number': 853}, 'overall_precision': 0.5885714285714285, 'overall_recall': 0.6186186186186187, 'overall_f1': 0.6032210834553441, 'overall_accuracy': 0.8256189451022605}
			------------EPOCH 12---------------
Loss:  tensor(0.0269, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0232, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0045, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0077, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2826, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0345, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0325, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0106, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2236, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1418, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0461, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2673, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2047, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0469, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0594, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0626, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0243, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0482, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0035, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0604, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0736, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1160, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0405, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0299, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0492, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0049, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0269, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0444, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0581, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0061, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3864, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3214, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0833, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0444, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0938, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1729, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0582, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0927, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5412621359223301, 'recall': 0.46555323590814196, 'f1': 0.500561167227834, 'number': 479}, 'P': {'precision': 0.6655773420479303, 'recall': 0.716295427901524, 'f1': 0.6900056465273856, 'number': 853}, 'overall_precision': 0.6270676691729323, 'overall_recall': 0.6261261261261262, 'overall_f1': 0.6265965439519158, 'overall_accuracy': 0.8268747757445282}
			------------EPOCH 13---------------
Loss:  tensor(0.0345, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0424, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0066, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0122, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2564, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0335, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0311, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0107, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1991, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1177, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0268, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3588, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2283, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1753, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0408, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0357, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0332, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0171, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0658, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0047, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0553, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0659, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1302, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0823, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0416, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1000, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0037, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0422, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0439, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0519, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0044, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3654, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2997, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0809, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0232, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0944, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1602, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0241, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0534, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5, 'recall': 0.46764091858037576, 'f1': 0.4832793959007551, 'number': 479}, 'P': {'precision': 0.6753393665158371, 'recall': 0.6998827667057445, 'f1': 0.687392055267703, 'number': 853}, 'overall_precision': 0.6163663663663663, 'overall_recall': 0.6163663663663663, 'overall_f1': 0.6163663663663663, 'overall_accuracy': 0.8211338356655903}
			------------EPOCH 14---------------
Loss:  tensor(0.0210, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0779, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0246, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0205, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2708, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0552, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0330, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0079, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2281, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1078, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0255, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3620, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2628, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1366, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0444, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0343, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0191, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0121, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0171, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0035, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0487, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0268, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1041, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0272, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0491, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0672, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0031, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0457, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0451, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0904, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0030, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4193, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3923, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0993, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0362, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0861, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1675, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0191, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0407, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.475, 'recall': 0.5156576200417536, 'f1': 0.4944944944944945, 'number': 479}, 'P': {'precision': 0.6504424778761062, 'recall': 0.6893317702227433, 'f1': 0.6693227091633467, 'number': 853}, 'overall_precision': 0.586376404494382, 'overall_recall': 0.6268768768768769, 'overall_f1': 0.6059506531204645, 'overall_accuracy': 0.8273830881473508}
			------------EPOCH 15---------------
Loss:  tensor(0.0107, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0212, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0033, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0048, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2491, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0823, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0574, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0054, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2072, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0943, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0232, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3729, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2697, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0600, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0301, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0219, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0106, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0184, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0605, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0251, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1328, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0212, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0456, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0478, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0129, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0496, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0682, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4463, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3179, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0903, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0245, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0592, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1673, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0326, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0538, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.43610223642172524, 'recall': 0.569937369519833, 'f1': 0.49411764705882355, 'number': 479}, 'P': {'precision': 0.6121343445287107, 'recall': 0.6623681125439624, 'f1': 0.6362612612612614, 'number': 853}, 'overall_precision': 0.5409941897998709, 'overall_recall': 0.6291291291291291, 'overall_f1': 0.5817424505380078, 'overall_accuracy': 0.8216421480684128}
			------------EPOCH 16---------------
Loss:  tensor(0.0224, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0239, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0111, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0095, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3029, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0394, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0223, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0034, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0534, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0474, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3613, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2204, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1099, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0451, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0782, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0674, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0074, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0138, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0514, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0267, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1219, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0313, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0434, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0416, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0177, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0209, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1084, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3816, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3102, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0782, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0171, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0552, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1946, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0486, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0350, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.49902534113060426, 'recall': 0.534446764091858, 'f1': 0.5161290322580645, 'number': 479}, 'P': {'precision': 0.5743639921722113, 'recall': 0.6881594372801876, 'f1': 0.6261333333333333, 'number': 853}, 'overall_precision': 0.549185667752443, 'overall_recall': 0.6328828828828829, 'overall_f1': 0.5880711545169166, 'overall_accuracy': 0.8225989714149026}
			------------EPOCH 17---------------
Loss:  tensor(0.0189, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0322, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0049, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0071, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2564, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0350, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0282, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0049, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2124, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0543, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0255, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3370, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2035, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0957, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0334, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0312, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0131, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0039, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0067, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0569, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0117, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0971, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0131, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0250, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0250, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0173, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0344, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0462, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3752, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2828, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0842, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0180, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0592, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1184, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0105, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0384, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5456790123456791, 'recall': 0.4613778705636743, 'f1': 0.5, 'number': 479}, 'P': {'precision': 0.5797773654916512, 'recall': 0.7327080890973037, 'f1': 0.6473329880890729, 'number': 853}, 'overall_precision': 0.5704652730950776, 'overall_recall': 0.6351351351351351, 'overall_f1': 0.6010657193605683, 'overall_accuracy': 0.82780169836144}
			------------EPOCH 18---------------
Loss:  tensor(0.0057, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0052, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0028, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2723, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0079, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0121, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1974, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0175, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0096, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3249, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2186, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1303, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0313, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0140, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0189, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0084, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0141, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0375, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0120, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0748, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0127, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0212, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0151, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0104, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0225, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0284, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3249, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2602, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0772, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0152, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0701, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1401, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0091, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0264, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4967032967032967, 'recall': 0.4718162839248434, 'f1': 0.4839400428265525, 'number': 479}, 'P': {'precision': 0.6549222797927461, 'recall': 0.7409144196951934, 'f1': 0.6952695269526953, 'number': 853}, 'overall_precision': 0.604225352112676, 'overall_recall': 0.6441441441441441, 'overall_f1': 0.623546511627907, 'overall_accuracy': 0.8331838296854444}
			------------EPOCH 19---------------
Loss:  tensor(0.0041, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0040, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1916, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0132, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1519, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0160, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0071, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2687, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1731, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1167, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0321, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0155, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0125, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0031, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0058, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0429, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0095, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1107, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0072, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0094, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0125, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0054, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0101, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0474, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3235, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2194, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0708, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0108, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0960, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1213, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0096, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0303, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.43934426229508194, 'recall': 0.5594989561586639, 'f1': 0.4921946740128558, 'number': 479}, 'P': {'precision': 0.6948130277442702, 'recall': 0.675263774912075, 'f1': 0.6848989298454221, 'number': 853}, 'overall_precision': 0.5865184155663655, 'overall_recall': 0.6336336336336337, 'overall_f1': 0.6091663659328761, 'overall_accuracy': 0.8232268867360364}
			------------EPOCH 20---------------
Loss:  tensor(0.0033, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0034, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3123, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0046, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0163, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2278, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0128, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0031, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3812, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2068, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0936, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0303, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0066, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0062, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0048, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0345, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0086, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1071, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0090, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0132, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0178, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0108, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0124, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0507, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3279, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2834, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0819, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0162, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0405, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1808, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0081, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0504, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4793233082706767, 'recall': 0.5323590814196242, 'f1': 0.5044510385756676, 'number': 479}, 'P': {'precision': 0.6456776947705443, 'recall': 0.7092614302461899, 'f1': 0.6759776536312849, 'number': 853}, 'overall_precision': 0.585432266848196, 'overall_recall': 0.6456456456456456, 'overall_f1': 0.6140664048554088, 'overall_accuracy': 0.8313897859107762}
			------------EPOCH 21---------------
Loss:  tensor(0.0026, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1702, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0033, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0090, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1536, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0089, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0032, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2508, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1663, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0894, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0236, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0077, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0086, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0030, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0325, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0084, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0735, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0045, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0086, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0075, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0060, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0077, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0416, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2735, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1962, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0640, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0136, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0312, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1088, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0058, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0376, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.46557971014492755, 'recall': 0.5365344467640919, 'f1': 0.498545101842871, 'number': 479}, 'P': {'precision': 0.6550598476605005, 'recall': 0.7057444314185228, 'f1': 0.6794582392776524, 'number': 853}, 'overall_precision': 0.583956492182189, 'overall_recall': 0.6448948948948949, 'overall_f1': 0.612914734213343, 'overall_accuracy': 0.8337818442770003}
			------------EPOCH 22---------------
Loss:  tensor(0.0018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1415, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1280, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0064, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2163, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1247, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1071, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0130, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0078, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0225, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0039, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0557, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0072, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0059, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0040, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0070, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0154, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2246, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1504, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0532, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0124, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0594, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0691, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0243, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4826254826254826, 'recall': 0.5219206680584552, 'f1': 0.5015045135406219, 'number': 479}, 'P': {'precision': 0.6414087513340448, 'recall': 0.7045720984759671, 'f1': 0.6715083798882681, 'number': 853}, 'overall_precision': 0.5848797250859107, 'overall_recall': 0.6388888888888888, 'overall_f1': 0.6106925008970219, 'overall_accuracy': 0.831569190288243}
			------------EPOCH 23---------------
Loss:  tensor(0.0015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1185, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0028, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1049, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0054, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1856, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1065, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0839, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0082, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0025, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0025, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0199, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0035, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0465, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0037, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0066, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0044, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0034, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0072, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0122, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2113, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1558, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0547, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0109, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0422, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0701, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0046, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0236, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.48330058939096265, 'recall': 0.5135699373695198, 'f1': 0.4979757085020243, 'number': 479}, 'P': {'precision': 0.6412698412698413, 'recall': 0.7104337631887456, 'f1': 0.6740823136818688, 'number': 853}, 'overall_precision': 0.5859697386519945, 'overall_recall': 0.6396396396396397, 'overall_f1': 0.6116295764536971, 'overall_accuracy': 0.833094127496711}
			------------EPOCH 24---------------
Loss:  tensor(0.0014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1030, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0963, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0052, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1706, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1064, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1095, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0098, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0198, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0029, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0489, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0025, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0061, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0036, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0029, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0063, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0098, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1763, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1227, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0422, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0104, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0645, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0589, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0257, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4205298013245033, 'recall': 0.5302713987473904, 'f1': 0.4690674053554939, 'number': 479}, 'P': {'precision': 0.6412803532008831, 'recall': 0.6811254396248535, 'f1': 0.6606026151222286, 'number': 853}, 'overall_precision': 0.5529801324503312, 'overall_recall': 0.6268768768768769, 'overall_f1': 0.5876143560872624, 'overall_accuracy': 0.8283997129529961}
			------------EPOCH 25---------------
Loss:  tensor(0.0016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1185, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0032, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0997, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0055, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1834, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1089, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0844, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0073, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0030, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0174, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0030, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0552, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0054, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0045, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0034, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0062, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0222, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2905, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2041, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0639, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0117, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0336, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0914, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0047, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0235, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.43859649122807015, 'recall': 0.5741127348643006, 'f1': 0.49728752260397835, 'number': 479}, 'P': {'precision': 0.6735870818915801, 'recall': 0.6846424384525205, 'f1': 0.6790697674418604, 'number': 853}, 'overall_precision': 0.5749665327978581, 'overall_recall': 0.6448948948948949, 'overall_f1': 0.607926397735315, 'overall_accuracy': 0.8280708049276402}
			------------EPOCH 26---------------
Loss:  tensor(0.0014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1513, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0055, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1321, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0058, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2500, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1313, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0788, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0131, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0049, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0025, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0324, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0777, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0034, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0047, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0053, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0057, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0203, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2229, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1533, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0467, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0127, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0270, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0951, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.4093e-05, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0044, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0248, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.535796766743649, 'recall': 0.48434237995824636, 'f1': 0.5087719298245613, 'number': 479}, 'P': {'precision': 0.639344262295082, 'recall': 0.731535756154748, 'f1': 0.68234007654456, 'number': 853}, 'overall_precision': 0.6075230660042583, 'overall_recall': 0.6426426426426426, 'overall_f1': 0.6245895658518789, 'overall_accuracy': 0.829476139217797}
			------------EPOCH 27---------------
Loss:  tensor(0.0015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1328, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0039, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1119, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0061, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1971, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1078, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0233, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0026, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0267, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0029, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0595, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0057, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0075, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2258, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1439, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0608, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0106, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0597, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0624, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0041, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0241, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5219206680584552, 'recall': 0.5219206680584552, 'f1': 0.5219206680584552, 'number': 479}, 'P': {'precision': 0.597029702970297, 'recall': 0.7069167643610785, 'f1': 0.6473429951690821, 'number': 853}, 'overall_precision': 0.5728676964405641, 'overall_recall': 0.6403903903903904, 'overall_f1': 0.6047500886210563, 'overall_accuracy': 0.8278315990910178}
			------------EPOCH 28---------------
Loss:  tensor(0.0016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1268, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0040, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1358, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0045, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1995, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1301, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0940, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0159, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0032, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0167, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0042, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0355, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0032, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0054, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0145, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2157, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1418, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0525, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0098, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0529, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0708, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0035, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0215, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.46964285714285714, 'recall': 0.5490605427974948, 'f1': 0.5062560153994226, 'number': 479}, 'P': {'precision': 0.6496191512513602, 'recall': 0.6998827667057445, 'f1': 0.6738148984198645, 'number': 853}, 'overall_precision': 0.581473968897904, 'overall_recall': 0.6456456456456456, 'overall_f1': 0.6118818925649235, 'overall_accuracy': 0.8307618705896423}
			------------EPOCH 29---------------
Loss:  tensor(0.0011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0841, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0731, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0034, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1480, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0676, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0870, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0082, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0189, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0030, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0373, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0028, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0050, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0036, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1602, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0894, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0390, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0092, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0399, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0332, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.8979e-05, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0030, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0214, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.47920433996383366, 'recall': 0.5532359081419624, 'f1': 0.5135658914728682, 'number': 479}, 'P': {'precision': 0.6633221850613155, 'recall': 0.6975381008206331, 'f1': 0.68, 'number': 853}, 'overall_precision': 0.593103448275862, 'overall_recall': 0.6456456456456456, 'overall_f1': 0.6182602444284686, 'overall_accuracy': 0.8300741538093529}
			------------EPOCH 30---------------
Loss:  tensor(0.0009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0782, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0934, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0029, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1467, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0696, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0833, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0060, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0088, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0225, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0026, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0048, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0058, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1458, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0926, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0316, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0083, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0319, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0486, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(6.9253e-05, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0028, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0213, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5104602510460251, 'recall': 0.5093945720250522, 'f1': 0.509926854754441, 'number': 479}, 'P': {'precision': 0.6292929292929293, 'recall': 0.7303634232121923, 'f1': 0.6760716223548562, 'number': 853}, 'overall_precision': 0.590599455040872, 'overall_recall': 0.6509009009009009, 'overall_f1': 0.6192857142857144, 'overall_accuracy': 0.8301339552685085}


		-------------RUN 2-----------
			------------EPOCH 1---------------
Loss:  tensor(3.5702, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5792, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.3905, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.9840, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.9262, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2215, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.9441, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.1954, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.8360, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.6096, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.6210, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.7842, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5930, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5114, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5557, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5760, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.4700, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.3746, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.3313, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9790, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3429, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5219, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2799, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2412, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.0667, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2644, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5125, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.0403, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.9053, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.9874, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3830, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8466, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.9832, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.7989, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.9468, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6489, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5816, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.7550, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6876, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6364, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3320610687022901, 'recall': 0.25970149253731345, 'f1': 0.2914572864321608, 'number': 335}, 'P': {'precision': 0.5061058344640434, 'recall': 0.6045380875202593, 'f1': 0.55096011816839, 'number': 617}, 'overall_precision': 0.46046046046046046, 'overall_recall': 0.4831932773109244, 'overall_f1': 0.47155304971809325, 'overall_accuracy': 0.7535989985395368}
			------------EPOCH 2---------------
Loss:  tensor(1.7227, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2884, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.7406, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.8901, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0206, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1259, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5214, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1624, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5264, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5089, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.8557, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5633, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6146, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5083, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6828, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4744, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4485, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2856, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3037, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6504, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9216, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3070, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1840, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3387, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3097, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4268, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4142, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2576, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3468, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0040, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5573, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5521, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4750, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4071, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4468, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1704, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2194, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2640, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.37318840579710144, 'recall': 0.3074626865671642, 'f1': 0.337152209492635, 'number': 335}, 'P': {'precision': 0.4484304932735426, 'recall': 0.6482982171799028, 'f1': 0.5301524188204109, 'number': 617}, 'overall_precision': 0.4306506849315068, 'overall_recall': 0.5283613445378151, 'overall_f1': 0.4745283018867924, 'overall_accuracy': 0.7841435426663885}
			------------EPOCH 3---------------
Loss:  tensor(1.3725, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2491, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3505, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5389, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7853, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0895, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1250, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8167, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1645, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1483, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3589, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1779, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1581, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1970, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1639, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0726, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0569, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8761, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9529, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4922, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7324, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2388, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1131, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9879, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9125, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0442, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7244, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0356, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8545, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9473, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7624, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3990, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3087, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1969, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0960, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3503, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8287, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8593, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8577, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0068, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4876847290640394, 'recall': 0.2955223880597015, 'f1': 0.3680297397769517, 'number': 335}, 'P': {'precision': 0.5139977603583427, 'recall': 0.7439222042139384, 'f1': 0.6079470198675497, 'number': 617}, 'overall_precision': 0.5091240875912408, 'overall_recall': 0.5861344537815126, 'overall_f1': 0.5449218749999999, 'overall_accuracy': 0.8028374713123305}
			------------EPOCH 4---------------
Loss:  tensor(1.0656, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1997, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4144, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5216, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0675, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7572, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5620, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8363, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8630, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1092, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8997, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8274, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9630, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7523, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7918, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5511, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3633, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6103, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1798, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0733, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6900, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6486, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7327, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5374, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7999, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6191, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7368, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6138, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2914, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0474, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9867, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9116, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3158, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7124, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7758, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6337, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7708, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4796747967479675, 'recall': 0.3522388059701492, 'f1': 0.40619621342512907, 'number': 335}, 'P': {'precision': 0.5593824228028503, 'recall': 0.7633711507293355, 'f1': 0.6456477039067855, 'number': 617}, 'overall_precision': 0.5413602941176471, 'overall_recall': 0.6186974789915967, 'overall_f1': 0.5774509803921568, 'overall_accuracy': 0.8184435635301481}
			------------EPOCH 5---------------
Loss:  tensor(0.7751, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1480, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8168, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1868, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4280, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0606, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7493, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4482, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6379, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6284, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8630, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6318, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6204, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7831, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6658, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7410, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7157, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6492, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6296, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2623, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4923, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1456, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0546, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5170, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5432, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5882, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4656, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6553, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6327, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7955, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6314, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3272, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9119, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8976, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6400, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2158, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4871, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5606, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5207, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7736, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4091778202676864, 'recall': 0.6388059701492538, 'f1': 0.49883449883449876, 'number': 335}, 'P': {'precision': 0.6097946287519748, 'recall': 0.6256077795786061, 'f1': 0.6175999999999999, 'number': 617}, 'overall_precision': 0.5190311418685121, 'overall_recall': 0.6302521008403361, 'overall_f1': 0.5692599620493358, 'overall_accuracy': 0.786855831420822}
			------------EPOCH 6---------------
Loss:  tensor(0.5944, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0977, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6271, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0977, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3039, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0313, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4199, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3404, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3589, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8178, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5301, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4660, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5440, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5772, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4604, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5989, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5413, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3940, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1921, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3053, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1336, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0281, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3034, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4221, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3449, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3633, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5458, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5300, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1771, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7900, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7622, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1893, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2919, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3834, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4404, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6677, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4068825910931174, 'recall': 0.6, 'f1': 0.48492159227985526, 'number': 335}, 'P': {'precision': 0.6117850953206239, 'recall': 0.5721231766612642, 'f1': 0.5912897822445561, 'number': 617}, 'overall_precision': 0.5172735760971056, 'overall_recall': 0.5819327731092437, 'overall_f1': 0.5477014335145823, 'overall_accuracy': 0.7874817442103067}
			------------EPOCH 7---------------
Loss:  tensor(0.5261, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4243, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9743, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1630, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0211, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2597, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1831, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2453, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2506, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6377, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3920, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4149, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5197, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2979, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3366, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2847, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2209, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2664, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2662, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1497, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0491, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3677, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5084, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3220, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2369, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3705, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1972, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3568, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4158, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1235, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7146, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9228, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8168, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1714, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2196, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3150, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2413, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5430, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4305019305019305, 'recall': 0.6656716417910448, 'f1': 0.5228604923798358, 'number': 335}, 'P': {'precision': 0.6679174484052532, 'recall': 0.5769854132901134, 'f1': 0.6191304347826085, 'number': 617}, 'overall_precision': 0.5509039010466222, 'overall_recall': 0.6081932773109243, 'overall_f1': 0.5781328007988018, 'overall_accuracy': 0.7900688504068433}
			------------EPOCH 8---------------
Loss:  tensor(0.2342, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0590, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3313, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9486, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1394, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0208, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3468, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3120, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2371, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3296, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7173, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3454, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3030, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3644, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1846, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2126, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3684, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2264, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3467, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1247, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2269, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1325, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0261, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2137, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1733, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1765, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1841, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2258, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1427, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2847, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3713, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1222, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5091, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6247, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3143, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0808, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1301, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2066, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1683, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5119, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.434878587196468, 'recall': 0.5880597014925373, 'f1': 0.5, 'number': 335}, 'P': {'precision': 0.6290322580645161, 'recall': 0.6320907617504052, 'f1': 0.6305578011317705, 'number': 617}, 'overall_precision': 0.5470643056849953, 'overall_recall': 0.6165966386554622, 'overall_f1': 0.5797530864197531, 'overall_accuracy': 0.8140204464844565}
			------------EPOCH 9---------------
Loss:  tensor(0.1483, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0303, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1521, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6468, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0683, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0092, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0928, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1251, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0837, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0918, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4926, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1695, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2056, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2619, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0732, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0916, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1050, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0919, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0920, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0241, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1126, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0821, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0064, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1045, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1079, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1058, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1552, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0811, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2120, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2841, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0702, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3904, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4935, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2532, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0513, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0790, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0964, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0766, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4169, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.41284403669724773, 'recall': 0.5373134328358209, 'f1': 0.46692607003891057, 'number': 335}, 'P': {'precision': 0.6286594761171033, 'recall': 0.6612641815235009, 'f1': 0.6445497630331755, 'number': 617}, 'overall_precision': 0.5419354838709678, 'overall_recall': 0.6176470588235294, 'overall_f1': 0.577319587628866, 'overall_accuracy': 0.8126851658668892}
			------------EPOCH 10---------------
Loss:  tensor(0.0796, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0164, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0867, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5385, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0285, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0057, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0386, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1847, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0478, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0488, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3895, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0706, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1670, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1721, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0448, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0420, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0799, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0463, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0545, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0161, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0824, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0727, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0033, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0513, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0748, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0643, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1064, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1340, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0692, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1578, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2517, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0622, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2885, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4487, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1769, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0273, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0477, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0593, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0374, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3653, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4075, 'recall': 0.48656716417910445, 'f1': 0.44353741496598637, 'number': 335}, 'P': {'precision': 0.6215827338129496, 'recall': 0.700162074554295, 'f1': 0.6585365853658538, 'number': 617}, 'overall_precision': 0.54337899543379, 'overall_recall': 0.625, 'overall_f1': 0.5813385442110406, 'overall_accuracy': 0.8153557271020238}
			------------EPOCH 11---------------
Loss:  tensor(0.0471, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0139, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0641, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4952, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0189, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0034, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0302, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0739, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0420, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0315, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3380, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0523, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1437, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1675, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0399, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0215, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0618, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0229, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0298, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0068, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0595, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0548, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0378, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0340, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0416, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0835, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1107, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0695, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1460, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2415, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0539, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2397, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4338, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1826, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0073, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0292, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0330, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0147, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3484, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.40425531914893614, 'recall': 0.5104477611940299, 'f1': 0.45118733509234826, 'number': 335}, 'P': {'precision': 0.6, 'recall': 0.6758508914100486, 'f1': 0.635670731707317, 'number': 617}, 'overall_precision': 0.5259391771019678, 'overall_recall': 0.6176470588235294, 'overall_f1': 0.5681159420289854, 'overall_accuracy': 0.8210306697266847}
			------------EPOCH 12---------------
Loss:  tensor(0.0251, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0069, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0434, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4977, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0103, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0025, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0173, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0562, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0186, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0169, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3054, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0356, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1269, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1269, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0165, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0156, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0461, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0189, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0248, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0056, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0808, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0533, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0292, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0194, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0255, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0533, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0765, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0268, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1088, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2315, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0437, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2150, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4105, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1120, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0039, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0171, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0175, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0105, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3300, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.49843260188087773, 'recall': 0.4746268656716418, 'f1': 0.48623853211009177, 'number': 335}, 'P': {'precision': 0.5721997300944669, 'recall': 0.6871961102106969, 'f1': 0.6244477172312224, 'number': 617}, 'overall_precision': 0.55, 'overall_recall': 0.6123949579831933, 'overall_f1': 0.5795228628230616, 'overall_accuracy': 0.8255789693302733}
			------------EPOCH 13---------------
Loss:  tensor(0.0202, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0346, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4700, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0094, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0032, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0219, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0823, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0244, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0157, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3600, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0401, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1304, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1192, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0099, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0096, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0387, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0160, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0280, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0043, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0743, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0631, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0586, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0369, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0412, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0536, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0675, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0322, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1422, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0732, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2534, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4890, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1755, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0044, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0258, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0299, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0231, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3434, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.46075085324232085, 'recall': 0.40298507462686567, 'f1': 0.4299363057324841, 'number': 335}, 'P': {'precision': 0.5424354243542435, 'recall': 0.7147487844408428, 'f1': 0.6167832167832168, 'number': 617}, 'overall_precision': 0.5207956600361664, 'overall_recall': 0.6050420168067226, 'overall_f1': 0.5597667638483965, 'overall_accuracy': 0.8184018360108491}
			------------EPOCH 14---------------
Loss:  tensor(0.0496, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0080, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0758, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4955, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0307, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0025, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0202, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0596, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0268, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0224, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3669, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0462, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1678, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1485, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0172, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0109, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0504, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0168, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0238, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0606, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0559, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0190, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0181, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0313, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0448, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0608, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0329, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0976, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2220, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0493, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2165, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4104, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1400, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0128, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0160, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0106, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3309, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4106280193236715, 'recall': 0.5074626865671642, 'f1': 0.4539385847797063, 'number': 335}, 'P': {'precision': 0.5977175463623395, 'recall': 0.6790923824959482, 'f1': 0.6358118361153262, 'number': 617}, 'overall_precision': 0.5282511210762332, 'overall_recall': 0.6186974789915967, 'overall_f1': 0.5699080793420417, 'overall_accuracy': 0.8135614437721678}
			------------EPOCH 15---------------
Loss:  tensor(0.0202, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0030, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0310, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4332, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0124, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0346, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0586, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0142, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0257, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2490, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0320, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0851, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0916, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0093, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0181, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0400, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0263, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0312, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0050, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0731, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0773, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0277, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0131, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0187, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0399, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0763, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0177, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0938, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2137, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0455, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1672, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3916, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1316, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0112, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0242, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0209, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0102, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3165, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4234592445328032, 'recall': 0.6358208955223881, 'f1': 0.5083532219570406, 'number': 335}, 'P': {'precision': 0.6072607260726073, 'recall': 0.5964343598055105, 'f1': 0.6017988552739167, 'number': 617}, 'overall_precision': 0.5238954012623985, 'overall_recall': 0.6102941176470589, 'overall_f1': 0.5638039786511403, 'overall_accuracy': 0.8026288337158356}
			------------EPOCH 16---------------
Loss:  tensor(0.0175, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0357, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4726, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0114, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0039, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0300, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0869, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0189, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0142, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2771, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0240, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0955, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0981, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0089, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0095, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0235, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0146, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0144, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0429, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0549, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0213, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0337, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0360, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0558, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0548, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0292, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1416, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2529, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0395, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1809, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3897, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1192, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0070, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0078, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0122, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0048, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2812, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3938730853391685, 'recall': 0.5373134328358209, 'f1': 0.45454545454545453, 'number': 335}, 'P': {'precision': 0.5768667642752562, 'recall': 0.6385737439222042, 'f1': 0.6061538461538462, 'number': 617}, 'overall_precision': 0.5035087719298246, 'overall_recall': 0.6029411764705882, 'overall_f1': 0.548757170172084, 'overall_accuracy': 0.8134362612142708}
			------------EPOCH 17---------------
Loss:  tensor(0.0130, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0295, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3820, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0091, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0291, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0856, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0137, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0082, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2473, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0231, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0917, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0912, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0089, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0116, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0224, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0173, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0146, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0302, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0535, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0101, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0072, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0143, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0385, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0371, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0191, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0687, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1983, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0445, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1761, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3520, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1208, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0088, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0084, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0352, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0066, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2484, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3787878787878788, 'recall': 0.44776119402985076, 'f1': 0.4103967168262654, 'number': 335}, 'P': {'precision': 0.5731225296442688, 'recall': 0.7050243111831442, 'f1': 0.6322674418604651, 'number': 617}, 'overall_precision': 0.5064935064935064, 'overall_recall': 0.6144957983193278, 'overall_f1': 0.5552918841955387, 'overall_accuracy': 0.8176090131441686}
			------------EPOCH 18---------------
Loss:  tensor(0.0096, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0232, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3913, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0046, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0287, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0570, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0145, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0092, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2858, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0483, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1116, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0144, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0149, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0168, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0122, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0093, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0382, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0487, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0095, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0102, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0140, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0321, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0540, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0181, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2554, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0542, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2121, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4156, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1126, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0101, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0215, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0442, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0128, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3141, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.46174142480211083, 'recall': 0.5223880597014925, 'f1': 0.4901960784313726, 'number': 335}, 'P': {'precision': 0.6738131699846861, 'recall': 0.713128038897893, 'f1': 0.6929133858267718, 'number': 617}, 'overall_precision': 0.5959302325581395, 'overall_recall': 0.6460084033613446, 'overall_f1': 0.6199596774193549, 'overall_accuracy': 0.8179845608178594}
			------------EPOCH 19---------------
Loss:  tensor(0.0083, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0187, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3760, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0089, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0394, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0067, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0070, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2327, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0176, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1072, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1149, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0520, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0297, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0401, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0469, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0268, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0089, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0432, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0461, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0151, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0142, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0127, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0319, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0264, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0092, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0653, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1828, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0326, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1538, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3146, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1643, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0148, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0151, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0269, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0214, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2615, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4713216957605985, 'recall': 0.564179104477612, 'f1': 0.5135869565217391, 'number': 335}, 'P': {'precision': 0.6793048973143759, 'recall': 0.6969205834683955, 'f1': 0.688, 'number': 617}, 'overall_precision': 0.5986460348162476, 'overall_recall': 0.6502100840336135, 'overall_f1': 0.6233635448136959, 'overall_accuracy': 0.8217817650740663}
			------------EPOCH 20---------------
Loss:  tensor(0.0431, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0335, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3901, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0055, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0093, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0731, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0042, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0042, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1814, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0164, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0681, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0623, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0101, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0096, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0560, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0494, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0447, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0131, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0508, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0468, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0025, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0373, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0229, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0187, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0427, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0247, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0090, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0688, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1981, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0270, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1230, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3383, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1196, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0156, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0207, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0110, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0262, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2599, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.47643979057591623, 'recall': 0.5432835820895522, 'f1': 0.5076708507670851, 'number': 335}, 'P': {'precision': 0.6492753623188405, 'recall': 0.7260940032414911, 'f1': 0.685539403213466, 'number': 617}, 'overall_precision': 0.5876865671641791, 'overall_recall': 0.6617647058823529, 'overall_f1': 0.6225296442687746, 'overall_accuracy': 0.8161902774880033}
			------------EPOCH 21---------------
Loss:  tensor(0.0312, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0034, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0591, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3765, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0119, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0140, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0714, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0075, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0120, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2100, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0404, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0902, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0843, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0070, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0056, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0149, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0076, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0744, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0139, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0769, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0418, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0028, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1178, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0485, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0891, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0558, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0743, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0214, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0727, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1627, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0259, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1332, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2907, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1225, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0129, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0347, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0233, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2379, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.48417721518987344, 'recall': 0.45671641791044776, 'f1': 0.4700460829493087, 'number': 335}, 'P': {'precision': 0.6340482573726541, 'recall': 0.766612641815235, 'f1': 0.694057226705796, 'number': 617}, 'overall_precision': 0.5894538606403014, 'overall_recall': 0.657563025210084, 'overall_f1': 0.621648460774578, 'overall_accuracy': 0.8069267682036303}
			------------EPOCH 22---------------
Loss:  tensor(0.0323, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0109, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1069, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3659, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0209, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0025, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0873, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1066, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0875, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0948, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1929, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0240, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0676, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0699, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0092, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0053, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0137, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0097, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0105, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0056, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0279, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0439, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0125, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0705, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0475, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0606, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1138, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1307, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0438, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1215, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2098, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0837, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2092, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3251, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1574, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0441, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0431, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0516, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0290, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2060, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4866666666666667, 'recall': 0.43582089552238806, 'f1': 0.4598425196850394, 'number': 335}, 'P': {'precision': 0.625974025974026, 'recall': 0.7811993517017828, 'f1': 0.6950252343186734, 'number': 617}, 'overall_precision': 0.5869158878504673, 'overall_recall': 0.6596638655462185, 'overall_f1': 0.6211671612265084, 'overall_accuracy': 0.8203630294179011}
			------------EPOCH 23---------------
Loss:  tensor(0.0239, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0460, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3032, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0082, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0208, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0538, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0465, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0472, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1960, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0516, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0965, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0906, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0375, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0355, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0249, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0858, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0626, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0454, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0239, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0469, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0256, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0356, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0151, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0361, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1105, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0171, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0379, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1417, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0272, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1086, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2874, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1201, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0030, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0320, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0366, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0618, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2069, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4649859943977591, 'recall': 0.4955223880597015, 'f1': 0.47976878612716767, 'number': 335}, 'P': {'precision': 0.6091644204851752, 'recall': 0.7325769854132901, 'f1': 0.665194996320824, 'number': 617}, 'overall_precision': 0.5623293903548681, 'overall_recall': 0.6491596638655462, 'overall_f1': 0.6026328620185276, 'overall_accuracy': 0.8101815147089505}
			------------EPOCH 24---------------
Loss:  tensor(0.0304, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0221, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0415, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2875, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0050, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0205, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0288, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0387, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0276, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1838, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0182, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0748, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0798, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0123, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0083, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0223, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0215, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0161, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0028, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0230, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0320, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0081, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0055, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0081, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0295, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0179, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0095, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0328, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1422, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0166, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0906, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2611, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1802, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0050, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0139, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0211, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0343, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1855, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4153498871331828, 'recall': 0.5492537313432836, 'f1': 0.4730077120822622, 'number': 335}, 'P': {'precision': 0.6447166921898928, 'recall': 0.6823338735818476, 'f1': 0.662992125984252, 'number': 617}, 'overall_precision': 0.552007299270073, 'overall_recall': 0.6355042016806722, 'overall_f1': 0.5908203125, 'overall_accuracy': 0.8045065720842896}
			------------EPOCH 25---------------
Loss:  tensor(0.0078, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0286, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3042, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0477, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0335, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1458, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0217, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0201, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1874, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0235, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0751, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0672, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0197, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0226, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0062, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0067, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0141, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0296, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0156, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0124, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0176, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0488, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0662, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0330, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0429, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1555, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0251, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0989, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2553, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1123, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0287, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0522, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0079, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0216, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2306, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4768856447688564, 'recall': 0.5850746268656717, 'f1': 0.5254691689008043, 'number': 335}, 'P': {'precision': 0.6750392464678179, 'recall': 0.6969205834683955, 'f1': 0.6858054226475279, 'number': 617}, 'overall_precision': 0.5973282442748091, 'overall_recall': 0.657563025210084, 'overall_f1': 0.6259999999999999, 'overall_accuracy': 0.8277488003338201}
			------------EPOCH 26---------------
Loss:  tensor(0.0077, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0184, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2429, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0049, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0278, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0113, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0140, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1581, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0042, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0435, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0530, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0119, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0143, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0457, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0357, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0213, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0177, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0226, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0376, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0161, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0168, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0309, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0644, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0073, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0372, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0383, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1398, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0148, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0785, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2690, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0953, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0225, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0227, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0048, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0079, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1655, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4631578947368421, 'recall': 0.5253731343283582, 'f1': 0.49230769230769234, 'number': 335}, 'P': {'precision': 0.6279761904761905, 'recall': 0.6839546191247974, 'f1': 0.6547711404189294, 'number': 617}, 'overall_precision': 0.5684410646387833, 'overall_recall': 0.6281512605042017, 'overall_f1': 0.5968063872255489, 'overall_accuracy': 0.8198622991863134}
			------------EPOCH 27---------------
Loss:  tensor(0.0137, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0062, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0301, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2308, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0216, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0082, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0425, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0134, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0114, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1374, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0166, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0357, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0332, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0127, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0067, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0072, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0083, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0156, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0255, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0076, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0067, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0118, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0341, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0185, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0049, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0315, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1219, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0159, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0680, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2155, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1491, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0109, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0043, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1463, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.45430809399477806, 'recall': 0.5194029850746269, 'f1': 0.4846796657381615, 'number': 335}, 'P': {'precision': 0.6508172362555721, 'recall': 0.7098865478119936, 'f1': 0.6790697674418605, 'number': 617}, 'overall_precision': 0.5795454545454546, 'overall_recall': 0.6428571428571429, 'overall_f1': 0.6095617529880478, 'overall_accuracy': 0.8277905278531191}
			------------EPOCH 28---------------
Loss:  tensor(0.0062, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0112, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0491, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0043, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1117, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0029, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0434, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0446, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0031, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0054, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0028, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0150, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0205, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0055, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0040, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0040, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0285, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0165, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0923, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0078, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0447, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1819, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0864, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1228, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.46272493573264784, 'recall': 0.5373134328358209, 'f1': 0.49723756906077354, 'number': 335}, 'P': {'precision': 0.6481203007518797, 'recall': 0.6985413290113452, 'f1': 0.6723868954758191, 'number': 617}, 'overall_precision': 0.579696394686907, 'overall_recall': 0.6418067226890757, 'overall_f1': 0.609172482552343, 'overall_accuracy': 0.8239933235969121}
			------------EPOCH 29---------------
Loss:  tensor(0.0026, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0101, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1780, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0439, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1058, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0272, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0211, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0093, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0166, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0053, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0025, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0033, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0346, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0139, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0098, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0399, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1769, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1029, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1042, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.46365914786967416, 'recall': 0.5522388059701493, 'f1': 0.5040871934604905, 'number': 335}, 'P': {'precision': 0.6491499227202473, 'recall': 0.6807131280388979, 'f1': 0.6645569620253164, 'number': 617}, 'overall_precision': 0.5783938814531548, 'overall_recall': 0.6355042016806722, 'overall_f1': 0.6056056056056055, 'overall_accuracy': 0.8189860212810348}
			------------EPOCH 30---------------
Loss:  tensor(0.0021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0092, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1546, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0464, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0956, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0291, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0190, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0093, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0184, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0046, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0029, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0350, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0151, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0952, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0062, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0310, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1685, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0956, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1220, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.46629213483146065, 'recall': 0.4955223880597015, 'f1': 0.4804630969609262, 'number': 335}, 'P': {'precision': 0.6225352112676056, 'recall': 0.7163695299837926, 'f1': 0.6661642803315749, 'number': 617}, 'overall_precision': 0.5703564727954972, 'overall_recall': 0.6386554621848739, 'overall_f1': 0.6025768087215063, 'overall_accuracy': 0.8306279991654496}


		-------------RUN 3-----------
			------------EPOCH 1---------------
Loss:  tensor(3.4704, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.4179, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.7564, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.8233, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3795, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.1009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.0988, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.7899, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.7995, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.8799, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.7476, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3799, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.8532, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.4524, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.4779, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2149, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.3195, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.3872, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2990, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4138, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2853, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.1723, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4805, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.0219, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.1030, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.0132, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6964, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.0681, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.9893, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1100, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8415, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6405, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.9845, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.9296, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1770, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.7543, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2129, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4860, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.7444, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3007518796992481, 'recall': 0.08771929824561403, 'f1': 0.1358234295415959, 'number': 456}, 'P': {'precision': 0.449119373776908, 'recall': 0.6428571428571429, 'f1': 0.5288018433179724, 'number': 714}, 'overall_precision': 0.43203463203463205, 'overall_recall': 0.4264957264957265, 'overall_f1': 0.42924731182795695, 'overall_accuracy': 0.7333152744387634}
			------------EPOCH 2---------------
Loss:  tensor(1.5962, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.7495, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6442, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.7186, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2217, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1146, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2801, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5890, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5473, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.7533, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6492, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2204, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0686, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4136, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5735, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6215, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3542, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5303, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3249, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2594, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3280, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4488, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2959, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3432, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4795, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3957, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4920, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3992, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3964, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8201, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5206, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1699, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3237, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4387, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0732, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7827, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2029, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8934, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3574, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2844, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3829787234042553, 'recall': 0.35526315789473684, 'f1': 0.36860068259385664, 'number': 456}, 'P': {'precision': 0.5638179800221975, 'recall': 0.711484593837535, 'f1': 0.6291021671826625, 'number': 714}, 'overall_precision': 0.5060422960725075, 'overall_recall': 0.5726495726495726, 'overall_f1': 0.5372894947874899, 'overall_accuracy': 0.8071987268479328}
			------------EPOCH 3---------------
Loss:  tensor(1.0647, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2196, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2283, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3824, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1300, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8713, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8493, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1053, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0845, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1713, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1190, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1497, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7332, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0261, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0509, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2336, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9124, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1591, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2048, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8699, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1169, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2176, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0028, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0125, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9485, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3412, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9484, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9994, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5549, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3796, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8683, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9344, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0820, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0447, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4903, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7734, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5697, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2727, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9323, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4843462246777164, 'recall': 0.5767543859649122, 'f1': 0.5265265265265264, 'number': 456}, 'P': {'precision': 0.6499356499356499, 'recall': 0.7072829131652661, 'f1': 0.6773977196512407, 'number': 714}, 'overall_precision': 0.5818181818181818, 'overall_recall': 0.6564102564102564, 'overall_f1': 0.616867469879518, 'overall_accuracy': 0.840111062201605}
			------------EPOCH 4---------------
Loss:  tensor(0.7493, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8602, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9484, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1338, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0686, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6787, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5861, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7648, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7484, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7582, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7422, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0931, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5440, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8167, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7106, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6542, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9296, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7833, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1675, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5857, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9568, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1806, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7984, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6834, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6720, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2666, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6537, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7593, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3964, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2680, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5764, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7148, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8785, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0283, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3143, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5388, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3979, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2183, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7201, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.46256239600665555, 'recall': 0.6096491228070176, 'f1': 0.5260170293282875, 'number': 456}, 'P': {'precision': 0.6002475247524752, 'recall': 0.6792717086834734, 'f1': 0.6373193166885677, 'number': 714}, 'overall_precision': 0.5415188076650106, 'overall_recall': 0.6521367521367522, 'overall_f1': 0.5917022101589763, 'overall_accuracy': 0.8338130227203467}
			------------EPOCH 5---------------
Loss:  tensor(0.5365, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6470, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7859, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9638, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0465, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5138, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3709, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5263, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5388, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4030, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0636, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4162, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6330, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5656, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0326, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4802, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6878, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4904, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0944, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5282, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9668, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1624, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7928, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8322, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6030, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2352, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5266, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7668, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3800, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2159, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4538, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5099, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7903, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0225, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2585, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5500, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3093, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2130, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8455, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.48478260869565215, 'recall': 0.48903508771929827, 'f1': 0.48689956331877726, 'number': 456}, 'P': {'precision': 0.6304093567251462, 'recall': 0.7549019607843137, 'f1': 0.687061822817081, 'number': 714}, 'overall_precision': 0.5794676806083651, 'overall_recall': 0.6512820512820513, 'overall_f1': 0.6132796780684105, 'overall_accuracy': 0.8355399045135949}
			------------EPOCH 6---------------
Loss:  tensor(0.4190, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6112, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6772, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8940, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0299, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4356, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3181, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4580, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3503, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3995, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3356, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0411, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3509, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4304, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4466, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7141, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3091, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6269, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4115, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0859, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2963, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7518, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1382, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5211, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5102, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2991, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1674, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4807, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4515, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2504, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1658, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2579, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3226, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6639, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0222, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1591, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3376, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2985, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1691, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5096, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.48747591522157996, 'recall': 0.5548245614035088, 'f1': 0.518974358974359, 'number': 456}, 'P': {'precision': 0.5739030023094688, 'recall': 0.696078431372549, 'f1': 0.6291139240506328, 'number': 714}, 'overall_precision': 0.5415162454873647, 'overall_recall': 0.6410256410256411, 'overall_f1': 0.5870841487279843, 'overall_accuracy': 0.8443436156164291}
			------------EPOCH 7---------------
Loss:  tensor(0.2286, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3563, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5772, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7497, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0280, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3586, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2288, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3197, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2136, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2084, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1660, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0289, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1986, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2215, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1946, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7192, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1725, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5987, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2101, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0282, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2122, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5554, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0950, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3880, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2713, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1917, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0832, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1840, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1931, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1666, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1879, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5705, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0083, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1511, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2179, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1321, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1469, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4835, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.46601941747572817, 'recall': 0.631578947368421, 'f1': 0.5363128491620112, 'number': 456}, 'P': {'precision': 0.6318681318681318, 'recall': 0.6442577030812325, 'f1': 0.6380027739251041, 'number': 714}, 'overall_precision': 0.5557206537890045, 'overall_recall': 0.6393162393162393, 'overall_f1': 0.5945945945945946, 'overall_accuracy': 0.8259912640097518}
			------------EPOCH 8---------------
Loss:  tensor(0.2654, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3215, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5149, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6705, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0339, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3327, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1853, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1041, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1309, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1183, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1356, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0246, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1502, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1480, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1778, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6053, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1287, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4773, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1304, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0161, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1423, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4354, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0890, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2964, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2263, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1807, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0838, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1723, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2822, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1592, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0742, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1113, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1185, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4334, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0522, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1446, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0893, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1377, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4000, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5068807339449541, 'recall': 0.48464912280701755, 'f1': 0.4955156950672646, 'number': 456}, 'P': {'precision': 0.6083608360836084, 'recall': 0.7745098039215687, 'f1': 0.6814540973505854, 'number': 714}, 'overall_precision': 0.5754646840148699, 'overall_recall': 0.6615384615384615, 'overall_f1': 0.6155069582504971, 'overall_accuracy': 0.8369281820336573}
			------------EPOCH 9---------------
Loss:  tensor(0.1469, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3343, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4308, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6170, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0059, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2985, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0782, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0791, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1133, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0917, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1130, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0199, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0940, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1715, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0708, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5530, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0754, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4322, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1096, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0089, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1268, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3614, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0736, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2525, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2590, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1420, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0580, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0839, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2056, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1060, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0572, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0817, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0869, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3269, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0063, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0422, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0588, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1195, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3073, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5175600739371534, 'recall': 0.6140350877192983, 'f1': 0.5616850551654965, 'number': 456}, 'P': {'precision': 0.5963636363636363, 'recall': 0.6890756302521008, 'f1': 0.6393762183235867, 'number': 714}, 'overall_precision': 0.5651537335285505, 'overall_recall': 0.6598290598290598, 'overall_f1': 0.6088328075709779, 'overall_accuracy': 0.8441743134798361}
			------------EPOCH 10---------------
Loss:  tensor(0.0760, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2135, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3759, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4906, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0032, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2396, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0814, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0273, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0464, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0547, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0834, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0102, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0632, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0880, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0464, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4126, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0449, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3578, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0884, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0060, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0440, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3067, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0595, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2147, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1228, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0513, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0387, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0734, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1595, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0761, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0498, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0535, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0494, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2789, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0209, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0743, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0279, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1037, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2755, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5093945720250522, 'recall': 0.5350877192982456, 'f1': 0.5219251336898396, 'number': 456}, 'P': {'precision': 0.5767077267637178, 'recall': 0.7212885154061625, 'f1': 0.6409458618543871, 'number': 714}, 'overall_precision': 0.5532069970845481, 'overall_recall': 0.6487179487179487, 'overall_f1': 0.5971675845790716, 'overall_accuracy': 0.8412284563031186}
			------------EPOCH 11---------------
Loss:  tensor(0.0419, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1398, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3420, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4109, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0043, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2159, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0468, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0331, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0413, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0556, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0338, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0421, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0300, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0179, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3912, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0295, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3266, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0643, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0623, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2505, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0478, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1743, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1125, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0383, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0354, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0487, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1256, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0522, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0405, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0266, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0288, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2252, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0100, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0424, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0181, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0922, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2260, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5338983050847458, 'recall': 0.5526315789473685, 'f1': 0.5431034482758621, 'number': 456}, 'P': {'precision': 0.5929705215419501, 'recall': 0.7324929971988795, 'f1': 0.6553884711779449, 'number': 714}, 'overall_precision': 0.5723781388478582, 'overall_recall': 0.6623931623931624, 'overall_f1': 0.6141045958795562, 'overall_accuracy': 0.8415670605763045}
			------------EPOCH 12---------------
Loss:  tensor(0.0272, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1151, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3059, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3671, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1991, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0604, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0137, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0294, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0474, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0227, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0033, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0388, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0263, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0204, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4517, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0256, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3345, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0529, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0230, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2284, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0531, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1196, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0868, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0116, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0320, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0545, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0752, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0523, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0398, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0254, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0290, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2510, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0119, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0378, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0167, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1045, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2455, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4870775347912525, 'recall': 0.5372807017543859, 'f1': 0.510948905109489, 'number': 456}, 'P': {'precision': 0.6093023255813953, 'recall': 0.7338935574229691, 'f1': 0.6658195679796696, 'number': 714}, 'overall_precision': 0.5641966250917094, 'overall_recall': 0.6572649572649573, 'overall_f1': 0.6071851559415713, 'overall_accuracy': 0.8421426878407205}
			------------EPOCH 13---------------
Loss:  tensor(0.0223, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1105, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2996, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3619, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1972, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1263, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0099, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0216, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0381, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0228, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0050, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0367, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0245, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0163, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5217, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0253, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3800, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0431, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0046, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0598, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2375, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0551, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1102, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0982, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0125, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0289, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0396, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0463, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0475, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0590, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0220, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0214, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2459, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0104, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0478, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0229, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1389, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3038, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5433884297520661, 'recall': 0.5767543859649122, 'f1': 0.5595744680851064, 'number': 456}, 'P': {'precision': 0.6746231155778895, 'recall': 0.7521008403361344, 'f1': 0.7112582781456953, 'number': 714}, 'overall_precision': 0.625, 'overall_recall': 0.6837606837606838, 'overall_f1': 0.653061224489796, 'overall_accuracy': 0.8503030508245014}
			------------EPOCH 14---------------
Loss:  tensor(0.0224, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1552, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4473, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4431, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2166, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0623, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0095, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0196, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0314, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0240, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0043, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0480, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0345, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0297, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5091, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0444, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4356, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0407, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0034, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0395, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3770, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0645, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1313, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1306, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0061, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0203, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0317, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0360, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0537, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0569, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0312, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0415, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2853, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0121, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0375, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0252, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1054, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2836, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5226860254083484, 'recall': 0.631578947368421, 'f1': 0.5719960278053624, 'number': 456}, 'P': {'precision': 0.6684563758389261, 'recall': 0.6974789915966386, 'f1': 0.682659355723098, 'number': 714}, 'overall_precision': 0.6064814814814815, 'overall_recall': 0.6717948717948717, 'overall_f1': 0.6374695863746959, 'overall_accuracy': 0.8461720786916331}
			------------EPOCH 15---------------
Loss:  tensor(0.0212, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1274, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4262, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4342, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2185, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0510, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0079, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0186, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0206, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0169, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0025, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0261, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0199, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0778, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4283, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0447, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3673, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0418, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0036, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0416, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0533, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1283, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1339, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0065, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0252, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0359, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0237, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0432, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0355, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0103, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0325, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0101, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0717, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0273, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0898, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2048, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.547576301615799, 'recall': 0.668859649122807, 'f1': 0.6021717670286278, 'number': 456}, 'P': {'precision': 0.7374631268436578, 'recall': 0.7002801120448179, 'f1': 0.7183908045977011, 'number': 714}, 'overall_precision': 0.6518218623481782, 'overall_recall': 0.688034188034188, 'overall_f1': 0.6694386694386695, 'overall_accuracy': 0.8486100294585718}
			------------EPOCH 16---------------
Loss:  tensor(0.0442, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1118, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3207, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3847, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1995, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0438, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0101, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0199, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0284, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0152, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0183, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0130, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0122, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3441, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0183, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2591, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0265, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0062, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0625, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2620, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0581, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1176, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1064, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0280, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0185, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0488, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0255, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0458, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0410, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0138, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0166, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1642, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0069, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0382, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0398, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0868, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1879, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5110294117647058, 'recall': 0.6096491228070176, 'f1': 0.5559999999999999, 'number': 456}, 'P': {'precision': 0.6859838274932615, 'recall': 0.7128851540616247, 'f1': 0.6991758241758242, 'number': 714}, 'overall_precision': 0.6119751166407466, 'overall_recall': 0.6726495726495727, 'overall_f1': 0.6408794788273616, 'overall_accuracy': 0.8336437205837538}
			------------EPOCH 17---------------
Loss:  tensor(0.0367, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1315, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3059, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3697, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1698, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0536, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1219, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0197, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0278, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0105, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0139, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0094, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0079, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3579, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0866, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3494, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0529, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0057, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0691, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2556, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0511, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0803, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0813, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0104, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0049, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0697, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0978, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0295, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0511, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0320, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0274, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1895, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0101, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0321, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0162, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0875, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2109, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.515828677839851, 'recall': 0.6074561403508771, 'f1': 0.5579053373615307, 'number': 456}, 'P': {'precision': 0.6628056628056628, 'recall': 0.7212885154061625, 'f1': 0.6908115358819584, 'number': 714}, 'overall_precision': 0.6027397260273972, 'overall_recall': 0.676923076923077, 'overall_f1': 0.6376811594202899, 'overall_accuracy': 0.8348288355399045}
			------------EPOCH 18---------------
Loss:  tensor(0.0442, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3186, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4543, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0056, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2025, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1025, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1760, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0899, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1069, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0441, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0050, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0184, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0203, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0160, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3188, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0486, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2984, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0651, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0040, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1329, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2171, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0412, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0876, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0839, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0444, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0689, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0979, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0932, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0359, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0363, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0242, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0317, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1820, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0083, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0209, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0265, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0733, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1764, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5271028037383177, 'recall': 0.618421052631579, 'f1': 0.5691220988900102, 'number': 456}, 'P': {'precision': 0.6609947643979057, 'recall': 0.7072829131652661, 'f1': 0.6833558863328822, 'number': 714}, 'overall_precision': 0.6058506543494996, 'overall_recall': 0.6726495726495727, 'overall_f1': 0.6375050627784529, 'overall_accuracy': 0.8385534825449498}
			------------EPOCH 19---------------
Loss:  tensor(0.0416, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0853, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2384, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2987, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1739, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0805, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0815, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0431, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0948, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0378, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0062, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0874, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0659, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0358, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3599, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0256, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2756, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0590, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0050, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0272, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1961, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0419, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0642, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0606, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0235, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1349, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1181, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0313, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0253, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0429, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0518, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0470, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1489, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0170, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0596, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0185, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0973, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2054, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5485148514851486, 'recall': 0.6074561403508771, 'f1': 0.5764828303850156, 'number': 456}, 'P': {'precision': 0.6583541147132169, 'recall': 0.7394957983193278, 'f1': 0.6965699208443272, 'number': 714}, 'overall_precision': 0.6159143075745983, 'overall_recall': 0.688034188034188, 'overall_f1': 0.6499798142914817, 'overall_accuracy': 0.8453594284359869}
			------------EPOCH 20---------------
Loss:  tensor(0.0249, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0903, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3187, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3249, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1667, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0700, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1100, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0134, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0583, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0228, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0152, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0129, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0079, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2889, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0256, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2406, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0368, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0560, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2338, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0491, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0753, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0498, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0182, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0106, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0251, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0283, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0224, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0204, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0176, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0155, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1041, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0138, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0156, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0132, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0661, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1428, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5404858299595142, 'recall': 0.5855263157894737, 'f1': 0.5621052631578949, 'number': 456}, 'P': {'precision': 0.6160493827160494, 'recall': 0.6988795518207283, 'f1': 0.6548556430446194, 'number': 714}, 'overall_precision': 0.5874233128834356, 'overall_recall': 0.6547008547008547, 'overall_f1': 0.6192400970088925, 'overall_accuracy': 0.8369620424609758}
			------------EPOCH 21---------------
Loss:  tensor(0.0176, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0655, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2367, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3044, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1287, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0444, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0053, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0106, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0200, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0096, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0055, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0026, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3388, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0113, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2710, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0200, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0298, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0339, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0642, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0652, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0056, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0246, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0164, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0253, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0352, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0062, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0088, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1358, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0084, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0674, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1634, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5466377440347071, 'recall': 0.5526315789473685, 'f1': 0.549618320610687, 'number': 456}, 'P': {'precision': 0.6071842410196987, 'recall': 0.7338935574229691, 'f1': 0.6645529486366518, 'number': 714}, 'overall_precision': 0.5861027190332326, 'overall_recall': 0.6632478632478632, 'overall_f1': 0.6222935044105853, 'overall_accuracy': 0.8376053905800291}
			------------EPOCH 22---------------
Loss:  tensor(0.0055, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0661, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2380, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2916, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1420, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0341, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0047, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0064, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0156, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0082, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0083, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0068, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0048, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2810, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0165, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2204, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0126, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0168, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1924, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0432, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0549, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0318, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0255, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0098, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0207, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0129, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0072, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0892, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0079, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0696, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1226, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.6105527638190955, 'recall': 0.5328947368421053, 'f1': 0.5690866510538641, 'number': 456}, 'P': {'precision': 0.6444444444444445, 'recall': 0.7717086834733894, 'f1': 0.7023581899298916, 'number': 714}, 'overall_precision': 0.6336791699920191, 'overall_recall': 0.6786324786324787, 'overall_f1': 0.6553858852661989, 'overall_accuracy': 0.8365218569058341}
			------------EPOCH 23---------------
Loss:  tensor(0.0062, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0575, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2295, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2817, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1204, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0235, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0123, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0076, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2527, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0073, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2053, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0126, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0308, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1538, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0321, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0440, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0271, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0099, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0089, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0175, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0097, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0057, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0799, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0061, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0578, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0994, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5929203539823009, 'recall': 0.5877192982456141, 'f1': 0.5903083700440529, 'number': 456}, 'P': {'precision': 0.6531862745098039, 'recall': 0.7464985994397759, 'f1': 0.6967320261437908, 'number': 714}, 'overall_precision': 0.6317034700315457, 'overall_recall': 0.6846153846153846, 'overall_f1': 0.6570959803117309, 'overall_accuracy': 0.842853756814411}
			------------EPOCH 24---------------
Loss:  tensor(0.0039, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0398, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1872, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2457, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0942, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0157, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0111, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0073, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3073, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0072, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2375, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0060, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0252, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1723, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0319, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0511, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0394, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0154, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0090, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0190, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0151, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0058, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0873, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0064, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0769, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1258, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5439560439560439, 'recall': 0.6513157894736842, 'f1': 0.5928143712574849, 'number': 456}, 'P': {'precision': 0.6479791395045632, 'recall': 0.696078431372549, 'f1': 0.6711681296421337, 'number': 714}, 'overall_precision': 0.6047220106626047, 'overall_recall': 0.6786324786324787, 'overall_f1': 0.6395489327426501, 'overall_accuracy': 0.8463413808282261}
			------------EPOCH 25---------------
Loss:  tensor(0.0038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0527, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2424, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2609, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1042, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0194, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0108, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0046, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0053, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2176, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1553, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0040, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0102, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1099, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0329, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0322, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0124, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0234, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0080, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0062, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0102, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0490, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0057, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0450, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0852, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5876068376068376, 'recall': 0.6030701754385965, 'f1': 0.5952380952380953, 'number': 456}, 'P': {'precision': 0.6662515566625156, 'recall': 0.7492997198879552, 'f1': 0.7053394858272907, 'number': 714}, 'overall_precision': 0.6372934697088907, 'overall_recall': 0.6923076923076923, 'overall_f1': 0.6636624334289225, 'overall_accuracy': 0.8455964514272171}
			------------EPOCH 26---------------
Loss:  tensor(0.0024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0411, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1591, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1980, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0790, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0253, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0101, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0042, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1942, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1428, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0233, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0987, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0239, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0215, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0099, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0108, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0074, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0048, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0073, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0048, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0495, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(9.0435e-05, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0055, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0468, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0879, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5748502994011976, 'recall': 0.631578947368421, 'f1': 0.6018808777429467, 'number': 456}, 'P': {'precision': 0.6484771573604061, 'recall': 0.7156862745098039, 'f1': 0.6804260985352862, 'number': 714}, 'overall_precision': 0.6198603568657874, 'overall_recall': 0.6829059829059829, 'overall_f1': 0.6498576657177715, 'overall_accuracy': 0.8447499407442522}
			------------EPOCH 27---------------
Loss:  tensor(0.0022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0358, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1408, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1747, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0671, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0590, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0100, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0036, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1783, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1377, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0205, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0861, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0227, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0234, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0095, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0134, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0070, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0048, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0059, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0047, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0344, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.8588e-05, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0053, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0398, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0599, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5748373101952278, 'recall': 0.581140350877193, 'f1': 0.5779716466739367, 'number': 456}, 'P': {'precision': 0.6403402187120292, 'recall': 0.7380952380952381, 'f1': 0.6857514638906963, 'number': 714}, 'overall_precision': 0.616822429906542, 'overall_recall': 0.676923076923077, 'overall_f1': 0.6454767726161369, 'overall_accuracy': 0.8405512477567467}
			------------EPOCH 28---------------
Loss:  tensor(0.0029, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0253, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1343, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1639, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0681, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0146, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0094, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0029, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1624, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1327, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0133, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0839, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0126, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0227, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0088, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0171, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0063, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0043, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0045, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0044, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0296, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.1526e-05, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0053, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0356, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0520, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5616740088105727, 'recall': 0.5592105263157895, 'f1': 0.5604395604395606, 'number': 456}, 'P': {'precision': 0.6264775413711584, 'recall': 0.742296918767507, 'f1': 0.6794871794871794, 'number': 714}, 'overall_precision': 0.6038461538461538, 'overall_recall': 0.6709401709401709, 'overall_f1': 0.6356275303643724, 'overall_accuracy': 0.838113296989808}
			------------EPOCH 29---------------
Loss:  tensor(0.0039, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0248, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1214, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1451, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0641, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0150, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0085, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2081, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1538, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0214, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0995, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0141, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0230, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0158, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0130, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0044, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0093, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0082, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0037, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0512, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(7.4110e-05, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0054, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0896, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1177, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5649606299212598, 'recall': 0.6293859649122807, 'f1': 0.5954356846473029, 'number': 456}, 'P': {'precision': 0.6688144329896907, 'recall': 0.726890756302521, 'f1': 0.6966442953020133, 'number': 714}, 'overall_precision': 0.6277258566978193, 'overall_recall': 0.6888888888888889, 'overall_f1': 0.6568867155664221, 'overall_accuracy': 0.8483730064673416}
			------------EPOCH 30---------------
Loss:  tensor(0.0040, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0323, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1457, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1856, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0519, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0136, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0080, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0025, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1989, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0029, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1697, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0043, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0217, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1499, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0263, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0518, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0618, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0154, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0050, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0123, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0110, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0722, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0053, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0446, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0736, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5767634854771784, 'recall': 0.6096491228070176, 'f1': 0.5927505330490405, 'number': 456}, 'P': {'precision': 0.6734434561626429, 'recall': 0.742296918767507, 'f1': 0.7061958694203864, 'number': 714}, 'overall_precision': 0.6367218282111899, 'overall_recall': 0.6905982905982906, 'overall_f1': 0.6625666256662567, 'overall_accuracy': 0.8487116107405276}


		-------------RUN 4-----------
			------------EPOCH 1---------------
Loss:  tensor(3.6477, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.3287, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.4644, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.1026, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2590, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.1994, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.9139, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.7612, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.7355, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2341, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3833, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5230, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3648, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.7177, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6142, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.8472, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.1669, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5324, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5254, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.4864, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1628, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3123, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2292, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2096, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.1634, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.1868, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.1638, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4967, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1591, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.9679, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.9652, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4432, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.9702, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.0889, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6747, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3537, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.0322, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.9148, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.9057, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.8067, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.16483516483516483, 'recall': 0.21479713603818615, 'f1': 0.18652849740932642, 'number': 419}, 'P': {'precision': 0.4194528875379939, 'recall': 0.5287356321839081, 'f1': 0.46779661016949153, 'number': 783}, 'overall_precision': 0.3287671232876712, 'overall_recall': 0.4193011647254576, 'overall_f1': 0.3685557586837294, 'overall_accuracy': 0.7492219656660978}
			------------EPOCH 2---------------
Loss:  tensor(1.6984, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.7220, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.8439, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1449, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7277, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.7027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6222, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5671, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.7984, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7322, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2381, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6202, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3207, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.7493, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4390, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0729, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5224, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.7141, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0780, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9262, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3470, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3898, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3781, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4614, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4387, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0363, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1092, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2993, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2906, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3177, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3675, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4687, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4438, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9157, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3919, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3316, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3473, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2468, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.291497975708502, 'recall': 0.5155131264916468, 'f1': 0.37241379310344824, 'number': 419}, 'P': {'precision': 0.49514563106796117, 'recall': 0.4559386973180077, 'f1': 0.47473404255319146, 'number': 783}, 'overall_precision': 0.39192886456908343, 'overall_recall': 0.4767054908485857, 'overall_f1': 0.43018018018018017, 'overall_accuracy': 0.7776662316367166}
			------------EPOCH 3---------------
Loss:  tensor(1.1625, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3941, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3218, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7942, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4941, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0653, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0941, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1122, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3590, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4544, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1710, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2408, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2964, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2155, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3278, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7433, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1070, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0265, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1361, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0513, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6411, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9297, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9928, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0377, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9902, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0112, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7386, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0788, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8367, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9476, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2383, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0855, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0659, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3798, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6603, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8785, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9134, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0417, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9263, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.2647058823529412, 'recall': 0.49403341288782815, 'f1': 0.3447127393838468, 'number': 419}, 'P': {'precision': 0.4840182648401826, 'recall': 0.541507024265645, 'f1': 0.5111512959614225, 'number': 783}, 'overall_precision': 0.38057901085645357, 'overall_recall': 0.5249584026622296, 'overall_f1': 0.4412587412587413, 'overall_accuracy': 0.8029314325870897}
			------------EPOCH 4---------------
Loss:  tensor(0.8174, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1650, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5459, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3672, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7294, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8160, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8350, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0536, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3361, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1310, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2956, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9834, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2835, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5950, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9109, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8332, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7169, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7803, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0401, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4563, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7596, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8781, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8791, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7826, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5708, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0408, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5971, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6335, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1751, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8898, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7745, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2767, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4652, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7273, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7944, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9446, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7406, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3914590747330961, 'recall': 0.5250596658711217, 'f1': 0.44852191641182465, 'number': 419}, 'P': {'precision': 0.6464285714285715, 'recall': 0.6934865900383141, 'f1': 0.6691312384473197, 'number': 783}, 'overall_precision': 0.5442225392296719, 'overall_recall': 0.6347753743760399, 'overall_f1': 0.5860215053763441, 'overall_accuracy': 0.8218719673392899}
			------------EPOCH 5---------------
Loss:  tensor(0.6399, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9713, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6732, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4222, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2690, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4965, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6417, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6118, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9795, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3285, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0850, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8982, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2320, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5906, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2648, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4134, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8145, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5899, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6218, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7033, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0258, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3928, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4716, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5449, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5339, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6290, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4687, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0302, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5682, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3988, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2130, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6995, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6125, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2362, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4250, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6884, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7514, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5656, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4681, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.33659730722154224, 'recall': 0.6563245823389021, 'f1': 0.4449838187702265, 'number': 419}, 'P': {'precision': 0.5769230769230769, 'recall': 0.5172413793103449, 'f1': 0.5454545454545454, 'number': 783}, 'overall_precision': 0.4476629361421988, 'overall_recall': 0.5657237936772047, 'overall_f1': 0.499816244027931, 'overall_accuracy': 0.7729812937121441}
			------------EPOCH 6---------------
Loss:  tensor(0.5382, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9469, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5952, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3780, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4183, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4319, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4201, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7475, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1923, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0742, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5793, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2169, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4980, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2285, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2968, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7477, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4554, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3521, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4892, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0166, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2436, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3098, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4203, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3480, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3693, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3062, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0210, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3874, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3690, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2155, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6946, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6942, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1745, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2383, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2688, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2647, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4260, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3432, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3574660633484163, 'recall': 0.37708830548926014, 'f1': 0.3670150987224158, 'number': 419}, 'P': {'precision': 0.5756097560975609, 'recall': 0.7535121328224776, 'f1': 0.6526548672566371, 'number': 783}, 'overall_precision': 0.5098841172460804, 'overall_recall': 0.6222961730449251, 'overall_f1': 0.5605095541401275, 'overall_accuracy': 0.8118662784860958}
			------------EPOCH 7---------------
Loss:  tensor(0.7145, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1697, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4981, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4092, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1234, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2613, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2843, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2573, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6820, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1909, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0467, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6896, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1915, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6624, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2593, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4583, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7132, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4075, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2347, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3458, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0157, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2361, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3443, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3308, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5861, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4140, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3762, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3517, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0165, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2601, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1608, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1113, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4629, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3742, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1491, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2540, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3915, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5618, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5374, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4027, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.41404805914972276, 'recall': 0.5346062052505967, 'f1': 0.4666666666666667, 'number': 419}, 'P': {'precision': 0.6574185765983113, 'recall': 0.6960408684546615, 'f1': 0.6761786600496279, 'number': 783}, 'overall_precision': 0.5613138686131387, 'overall_recall': 0.6397670549084858, 'overall_f1': 0.5979782270606531, 'overall_accuracy': 0.8151791988756149}
			------------EPOCH 8---------------
Loss:  tensor(0.1945, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6583, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2365, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1195, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1409, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3438, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3532, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3267, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6926, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1590, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0472, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4184, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1506, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4349, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1475, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1226, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5388, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2369, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1881, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2754, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0193, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1807, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2436, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2617, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2963, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2715, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2340, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2317, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0182, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2221, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1541, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4652, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3460, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0769, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1734, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1697, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1565, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2872, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2517, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.39625850340136054, 'recall': 0.5560859188544153, 'f1': 0.4627606752730884, 'number': 419}, 'P': {'precision': 0.6382211538461539, 'recall': 0.6781609195402298, 'f1': 0.6575851393188855, 'number': 783}, 'overall_precision': 0.5380281690140845, 'overall_recall': 0.6356073211314476, 'overall_f1': 0.5827612509534705, 'overall_accuracy': 0.8198976006425057}
			------------EPOCH 9---------------
Loss:  tensor(0.1327, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5771, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1538, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0847, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0600, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1638, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1221, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1174, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5158, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1162, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0347, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3081, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1407, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3173, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1522, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1431, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5151, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2258, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1516, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2669, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0075, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1339, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1674, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0824, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1317, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1797, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1265, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1343, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0135, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1640, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0996, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1086, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4464, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3057, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0778, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1166, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1207, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1675, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2310, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2533, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.49478079331941544, 'recall': 0.5656324582338902, 'f1': 0.5278396436525613, 'number': 419}, 'P': {'precision': 0.6871794871794872, 'recall': 0.6845466155810983, 'f1': 0.6858605246321177, 'number': 783}, 'overall_precision': 0.613979348689436, 'overall_recall': 0.6430948419301165, 'overall_f1': 0.6281999187322227, 'overall_accuracy': 0.8218719673392899}
			------------EPOCH 10---------------
Loss:  tensor(0.1809, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6182, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1236, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0781, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0475, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1784, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1031, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1086, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5094, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1076, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0100, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2669, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1317, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3160, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1436, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0812, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5036, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1556, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1165, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2201, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0091, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1502, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1700, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1082, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1601, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1552, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0965, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1241, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0098, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1244, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0610, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1036, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4117, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2589, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0540, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0698, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1147, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1194, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2070, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2207, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.39325842696629215, 'recall': 0.5847255369928401, 'f1': 0.47024952015355087, 'number': 419}, 'P': {'precision': 0.7176820208023774, 'recall': 0.6168582375478927, 'f1': 0.6634615384615384, 'number': 783}, 'overall_precision': 0.5617283950617284, 'overall_recall': 0.6056572379367721, 'overall_f1': 0.5828662930344275, 'overall_accuracy': 0.8129371214402837}
			------------EPOCH 11---------------
Loss:  tensor(0.1298, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5663, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0996, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0789, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0339, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1646, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1625, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1092, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3742, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0810, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0100, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2384, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1191, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1800, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1313, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1084, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4753, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1266, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0565, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1508, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0049, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1238, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1086, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0574, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1044, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1240, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1357, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0139, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1838, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1289, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1235, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3944, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2496, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0545, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0656, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0898, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1037, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1795, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1761, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3606789250353607, 'recall': 0.6085918854415274, 'f1': 0.45293072824156305, 'number': 419}, 'P': {'precision': 0.6769662921348315, 'recall': 0.6155810983397191, 'f1': 0.6448160535117057, 'number': 783}, 'overall_precision': 0.5193798449612403, 'overall_recall': 0.6131447587354409, 'overall_f1': 0.5623807706982068, 'overall_accuracy': 0.8111300739550915}
			------------EPOCH 12---------------
Loss:  tensor(0.0888, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0672, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0412, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0281, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0759, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0536, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0712, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0827, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0096, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2489, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0881, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2994, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1284, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1045, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4502, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1944, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1060, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2340, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0030, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0969, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0593, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0475, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0615, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0601, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0365, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0929, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0081, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0685, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0382, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0744, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3301, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1921, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0220, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0588, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0861, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1214, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1699, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1822, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3035190615835777, 'recall': 0.49403341288782815, 'f1': 0.3760217983651226, 'number': 419}, 'P': {'precision': 0.5223555070883316, 'recall': 0.611749680715198, 'f1': 0.563529411764706, 'number': 783}, 'overall_precision': 0.4290181363352095, 'overall_recall': 0.5707154742096506, 'overall_f1': 0.48982506247768653, 'overall_accuracy': 0.8066793829267477}
			------------EPOCH 13---------------
Loss:  tensor(0.0838, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4928, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0934, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0347, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0188, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0875, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0299, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0458, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2577, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0579, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1554, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0696, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1722, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1161, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0543, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3736, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0652, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0436, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1422, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0029, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0721, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0847, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0576, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0820, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0979, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0538, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1314, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0440, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0170, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0704, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2997, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1756, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0085, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0281, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0296, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0586, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1041, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1196, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3201267828843106, 'recall': 0.4821002386634845, 'f1': 0.38476190476190475, 'number': 419}, 'P': {'precision': 0.5780141843971631, 'recall': 0.6245210727969349, 'f1': 0.6003683241252302, 'number': 783}, 'overall_precision': 0.46784021665538256, 'overall_recall': 0.5748752079866889, 'overall_f1': 0.5158641284061217, 'overall_accuracy': 0.8094568818391727}
			------------EPOCH 14---------------
Loss:  tensor(0.0381, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4361, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0503, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0191, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0181, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0519, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0454, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0652, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2336, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0606, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0047, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1576, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0605, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1956, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1108, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0506, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3845, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0520, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0254, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0810, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0793, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0352, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0119, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0453, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0372, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0193, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0774, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0298, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0221, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0563, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3458, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1980, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0177, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0357, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0344, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0770, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1085, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1292, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5233160621761658, 'recall': 0.4821002386634845, 'f1': 0.5018633540372671, 'number': 419}, 'P': {'precision': 0.6116402116402117, 'recall': 0.7381864623243933, 'f1': 0.6689814814814815, 'number': 783}, 'overall_precision': 0.5860255447032306, 'overall_recall': 0.64891846921797, 'overall_f1': 0.6158705092775365, 'overall_accuracy': 0.8213700097045142}
			------------EPOCH 15---------------
Loss:  tensor(0.0368, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4869, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0215, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0066, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0044, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0347, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0106, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0111, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2061, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0474, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0036, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1223, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0637, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1374, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1064, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0414, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4162, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0322, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0297, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0889, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0645, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0384, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0239, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0374, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0346, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0145, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0435, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0189, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0147, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0766, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3150, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1899, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0044, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0180, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0097, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0366, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0632, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1133, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4702127659574468, 'recall': 0.5274463007159904, 'f1': 0.49718785151856015, 'number': 419}, 'P': {'precision': 0.6505190311418685, 'recall': 0.7203065134099617, 'f1': 0.6836363636363636, 'number': 783}, 'overall_precision': 0.587135377711294, 'overall_recall': 0.6530782029950083, 'overall_f1': 0.6183536825521858, 'overall_accuracy': 0.8258207007328582}
			------------EPOCH 16---------------
Loss:  tensor(0.0246, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3992, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0154, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0045, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0033, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0229, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0070, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0076, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2399, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0526, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0046, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1268, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0714, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1396, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0312, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3631, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0377, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0138, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0790, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0537, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0303, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0058, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0301, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0164, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0091, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0342, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0133, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0076, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0625, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2586, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1540, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0152, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0065, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0353, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0493, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1109, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.37083993660855785, 'recall': 0.5584725536992841, 'f1': 0.44571428571428573, 'number': 419}, 'P': {'precision': 0.6874172185430464, 'recall': 0.6628352490421456, 'f1': 0.6749024707412223, 'number': 783}, 'overall_precision': 0.5432900432900433, 'overall_recall': 0.6264559068219634, 'overall_f1': 0.5819165378670789, 'overall_accuracy': 0.8188267576883178}
			------------EPOCH 17---------------
Loss:  tensor(0.0277, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4353, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0151, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0028, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0032, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0091, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0042, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1685, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0364, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0915, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0532, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1159, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0894, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0735, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3254, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0142, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0109, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0675, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0555, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0247, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0050, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0270, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0119, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0083, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0560, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0100, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0040, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0550, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2978, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1477, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0124, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0044, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0334, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0329, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0734, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3639291465378422, 'recall': 0.5393794749403341, 'f1': 0.4346153846153846, 'number': 419}, 'P': {'precision': 0.6439024390243903, 'recall': 0.6743295019157088, 'f1': 0.6587648159700561, 'number': 783}, 'overall_precision': 0.5232477446217905, 'overall_recall': 0.627287853577371, 'overall_f1': 0.5705637533106319, 'overall_accuracy': 0.8183248000535421}
			------------EPOCH 18---------------
Loss:  tensor(0.0150, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3474, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0139, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0074, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0034, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0061, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1933, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0443, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1059, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0577, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1532, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0900, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0172, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3289, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0102, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0095, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0452, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0434, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0243, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0040, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0204, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0102, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0069, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0700, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0088, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0036, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0424, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2213, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1230, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0105, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0042, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0346, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0276, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0757, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4645161290322581, 'recall': 0.5155131264916468, 'f1': 0.48868778280542985, 'number': 419}, 'P': {'precision': 0.6347237880496054, 'recall': 0.719029374201788, 'f1': 0.674251497005988, 'number': 783}, 'overall_precision': 0.5761834319526628, 'overall_recall': 0.6480865224625624, 'overall_f1': 0.6100234925606891, 'overall_accuracy': 0.8287320550145568}
			------------EPOCH 19---------------
Loss:  tensor(0.0140, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3509, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0110, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0071, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0028, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0030, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1459, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0334, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0749, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0460, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0990, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0783, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0354, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2770, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0076, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0069, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0425, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0356, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0193, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0028, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0165, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0078, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0056, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0492, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0059, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0026, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0404, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1943, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0078, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0029, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0275, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0252, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0581, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.42023346303501946, 'recall': 0.5155131264916468, 'f1': 0.46302250803858525, 'number': 419}, 'P': {'precision': 0.6178128523111612, 'recall': 0.6998722860791826, 'f1': 0.6562874251497006, 'number': 783}, 'overall_precision': 0.5453247680228408, 'overall_recall': 0.6356073211314476, 'overall_f1': 0.587014982712255, 'overall_accuracy': 0.8276946759026872}
			------------EPOCH 20---------------
Loss:  tensor(0.0130, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3177, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0097, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0065, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1203, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0269, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0562, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0438, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1033, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0735, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0303, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2409, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0067, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0060, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0345, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0279, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0220, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0127, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0071, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0045, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0338, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0047, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0349, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1605, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0898, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0068, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0235, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0215, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0481, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.39191564147627417, 'recall': 0.5322195704057279, 'f1': 0.4514170040485829, 'number': 419}, 'P': {'precision': 0.6491442542787286, 'recall': 0.6781609195402298, 'f1': 0.6633354153653966, 'number': 783}, 'overall_precision': 0.5436193222782985, 'overall_recall': 0.627287853577371, 'overall_f1': 0.5824642719196601, 'overall_accuracy': 0.817153565572399}
			------------EPOCH 21---------------
Loss:  tensor(0.0124, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2964, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0087, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0061, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1030, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0241, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0440, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0408, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0962, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0724, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0340, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2468, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0053, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0057, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0317, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0197, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0209, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0082, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0062, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0036, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0347, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0034, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0326, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1563, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0891, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0061, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0222, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0160, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0521, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4089285714285714, 'recall': 0.5465393794749404, 'f1': 0.4678243105209397, 'number': 419}, 'P': {'precision': 0.6441495778045838, 'recall': 0.6819923371647509, 'f1': 0.662531017369727, 'number': 783}, 'overall_precision': 0.5493160547156227, 'overall_recall': 0.6347753743760399, 'overall_f1': 0.5889617908143574, 'overall_accuracy': 0.8177224508918114}
			------------EPOCH 22---------------
Loss:  tensor(0.0107, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2773, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0070, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0058, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0937, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0243, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0460, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0414, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0643, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0319, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2349, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0082, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0054, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0433, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0285, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0165, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0053, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0063, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0029, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0319, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0041, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0491, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2399, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1283, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0055, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0217, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0135, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0499, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.40932642487046633, 'recall': 0.5656324582338902, 'f1': 0.4749498997995992, 'number': 419}, 'P': {'precision': 0.6577608142493638, 'recall': 0.6602809706257982, 'f1': 0.6590184831102613, 'number': 783}, 'overall_precision': 0.5523809523809524, 'overall_recall': 0.627287853577371, 'overall_f1': 0.5874561745227892, 'overall_accuracy': 0.8145433858715658}
			------------EPOCH 23---------------
Loss:  tensor(0.0101, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2855, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0067, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0064, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0028, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2254, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0668, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1089, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0474, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0929, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0840, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0427, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2940, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0108, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0061, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0489, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0493, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0216, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0025, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0068, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0068, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0046, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0263, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0048, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0495, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2456, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1339, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0082, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0026, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0157, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0120, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0881, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3996138996138996, 'recall': 0.49403341288782815, 'f1': 0.4418356456776948, 'number': 419}, 'P': {'precision': 0.6526315789473685, 'recall': 0.7126436781609196, 'f1': 0.6813186813186813, 'number': 783}, 'overall_precision': 0.5571740713765477, 'overall_recall': 0.6364392678868552, 'overall_f1': 0.5941747572815533, 'overall_accuracy': 0.8262222668406787}
			------------EPOCH 24---------------
Loss:  tensor(0.0105, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2905, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0067, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0062, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1145, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0298, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0491, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0425, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1398, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0794, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0186, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2779, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0198, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0057, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0495, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0391, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0186, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0049, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0045, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0052, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0279, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0043, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0379, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1618, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0881, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0060, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0206, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0126, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0746, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.45372460496614, 'recall': 0.4797136038186158, 'f1': 0.4663573085846868, 'number': 419}, 'P': {'precision': 0.5557809330628803, 'recall': 0.6998722860791826, 'f1': 0.6195590729225552, 'number': 783}, 'overall_precision': 0.5241427571728482, 'overall_recall': 0.6231281198003328, 'overall_f1': 0.5693652603572786, 'overall_accuracy': 0.8257872368905398}
			------------EPOCH 25---------------
Loss:  tensor(0.0113, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3494, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0062, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0062, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0030, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1212, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0443, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0539, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0427, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0889, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0715, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0589, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2856, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0168, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0059, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0354, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0299, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0194, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0046, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0041, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0048, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0362, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0117, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0461, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1771, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1174, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0056, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0063, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0148, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0113, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0428, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.43339960238568587, 'recall': 0.5202863961813843, 'f1': 0.47288503253796094, 'number': 419}, 'P': {'precision': 0.6377314814814815, 'recall': 0.7037037037037037, 'f1': 0.6690953248330298, 'number': 783}, 'overall_precision': 0.562545720555962, 'overall_recall': 0.6397670549084858, 'overall_f1': 0.5986765278318411, 'overall_accuracy': 0.8277281397450055}
			------------EPOCH 26---------------
Loss:  tensor(0.0096, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2618, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0057, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1053, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0320, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0450, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0522, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1576, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0775, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0131, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2520, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0196, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0053, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0451, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0170, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0182, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0030, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0262, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0349, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1260, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0718, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0049, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0173, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0109, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0370, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4365942028985507, 'recall': 0.5751789976133651, 'f1': 0.49639546858908346, 'number': 419}, 'P': {'precision': 0.6817010309278351, 'recall': 0.6756066411238825, 'f1': 0.6786401539448365, 'number': 783}, 'overall_precision': 0.5798192771084337, 'overall_recall': 0.6405990016638935, 'overall_f1': 0.608695652173913, 'overall_accuracy': 0.820366094434963}
			------------EPOCH 27---------------
Loss:  tensor(0.0109, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2700, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0048, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0056, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0884, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0167, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0317, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0366, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0871, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0582, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0469, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2172, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0050, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0237, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0125, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0146, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0232, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0324, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1159, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0595, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0042, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0184, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0085, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0273, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3833634719710669, 'recall': 0.5059665871121718, 'f1': 0.43621399176954734, 'number': 419}, 'P': {'precision': 0.639618138424821, 'recall': 0.6845466155810983, 'f1': 0.6613201727328809, 'number': 783}, 'overall_precision': 0.5377426312005751, 'overall_recall': 0.6222961730449251, 'overall_f1': 0.5769379097570382, 'overall_accuracy': 0.8240471170899842}
			------------EPOCH 28---------------
Loss:  tensor(0.0075, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2220, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0092, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0053, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0613, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0092, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0243, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0389, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1244, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0461, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0215, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1555, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0049, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0159, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0131, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0200, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0232, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0273, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0911, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0506, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0041, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0142, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0072, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0217, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.43548387096774194, 'recall': 0.5155131264916468, 'f1': 0.47213114754098356, 'number': 419}, 'P': {'precision': 0.6284403669724771, 'recall': 0.6998722860791826, 'f1': 0.6622356495468278, 'number': 783}, 'overall_precision': 0.5584795321637427, 'overall_recall': 0.6356073211314476, 'overall_f1': 0.5945525291828794, 'overall_accuracy': 0.8292340126493324}
			------------EPOCH 29---------------
Loss:  tensor(0.0071, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0093, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0050, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0534, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0087, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0226, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0338, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0872, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0418, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0435, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1587, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0044, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0125, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0093, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0165, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0273, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0275, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0848, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0468, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0039, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0150, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0070, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0153, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.47608695652173916, 'recall': 0.522673031026253, 'f1': 0.49829351535836175, 'number': 419}, 'P': {'precision': 0.6047774158523345, 'recall': 0.7113665389527458, 'f1': 0.653755868544601, 'number': 783}, 'overall_precision': 0.5619116582186822, 'overall_recall': 0.6455906821963394, 'overall_f1': 0.6008517228029423, 'overall_accuracy': 0.8250844962018539}
			------------EPOCH 30---------------
Loss:  tensor(0.0066, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1763, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0093, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0052, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0519, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0068, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0221, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0372, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1269, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0385, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0176, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1477, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0043, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0167, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(8.6798e-05, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0057, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0144, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0285, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0288, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0792, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0371, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0033, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0203, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0083, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0220, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3814262023217247, 'recall': 0.548926014319809, 'f1': 0.45009784735812136, 'number': 419}, 'P': {'precision': 0.6442185514612452, 'recall': 0.6475095785440613, 'f1': 0.645859872611465, 'number': 783}, 'overall_precision': 0.5302158273381294, 'overall_recall': 0.6131447587354409, 'overall_f1': 0.5686728395061728, 'overall_accuracy': 0.8163838971990764}


		-------------RUN 5-----------
			------------EPOCH 1---------------
Loss:  tensor(3.4726, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.0502, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.2030, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.0044, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.9457, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4977, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3763, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6637, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.7984, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.6593, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.6003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.8789, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5245, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5035, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.3835, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.4724, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6681, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0344, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6361, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.1709, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2370, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.1248, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.0999, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5069, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2789, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.9521, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.8118, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.8324, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.8852, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0876, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.7978, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.8813, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.9130, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.7742, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7466, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.7096, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2864, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5931, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.7111, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.24778761061946902, 'recall': 0.12903225806451613, 'f1': 0.16969696969696968, 'number': 434}, 'P': {'precision': 0.33525456292026895, 'recall': 0.5124816446402349, 'f1': 0.4053426248548199, 'number': 681}, 'overall_precision': 0.3196527229676401, 'overall_recall': 0.3632286995515695, 'overall_f1': 0.34005037783375314, 'overall_accuracy': 0.7324930491195552}
			------------EPOCH 2---------------
Loss:  tensor(1.7177, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5868, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5255, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0947, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5200, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2831, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3077, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3546, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6955, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4289, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5081, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1062, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5185, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4785, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4681, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5179, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0208, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6263, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9465, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5727, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5764, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5528, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4030, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3749, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1867, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2420, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3698, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0621, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1633, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8081, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2028, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3267, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3749, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1195, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5143, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2558, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9253, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3865, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1240, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.32616487455197135, 'recall': 0.41935483870967744, 'f1': 0.36693548387096775, 'number': 434}, 'P': {'precision': 0.41161290322580646, 'recall': 0.4684287812041116, 'f1': 0.4381868131868132, 'number': 681}, 'overall_precision': 0.3758439609902476, 'overall_recall': 0.4493273542600897, 'overall_f1': 0.40931372549019607, 'overall_accuracy': 0.7736793327154773}
			------------EPOCH 3---------------
Loss:  tensor(1.2463, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2078, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0298, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7384, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0092, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2073, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2743, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2529, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2071, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9756, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9911, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8304, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1087, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0571, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0520, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1234, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6823, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3793, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6379, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0037, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0514, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2849, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0355, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0701, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1040, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9441, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1770, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7704, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8097, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5704, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8331, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9395, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0629, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7840, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4036, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9805, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6946, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2814, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7997, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.39092495636998253, 'recall': 0.5161290322580645, 'f1': 0.4448857994041707, 'number': 434}, 'P': {'precision': 0.511049723756906, 'recall': 0.5433186490455213, 'f1': 0.5266903914590748, 'number': 681}, 'overall_precision': 0.4579799537393986, 'overall_recall': 0.5327354260089686, 'overall_f1': 0.4925373134328358, 'overall_accuracy': 0.7982576459684894}
			------------EPOCH 4---------------
Loss:  tensor(0.9079, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9512, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7942, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4829, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7181, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1536, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2369, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1675, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8306, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7220, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7789, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6896, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7834, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7293, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7370, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7300, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5121, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3146, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5480, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0440, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6919, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0460, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5986, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8672, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0706, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8144, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3081, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5555, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6384, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3815, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7584, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9275, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6477, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3075, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8489, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4417, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6416, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.504297994269341, 'recall': 0.4055299539170507, 'f1': 0.4495530012771392, 'number': 434}, 'P': {'precision': 0.58, 'recall': 0.723935389133627, 'f1': 0.644023514043109, 'number': 681}, 'overall_precision': 0.5579649708090075, 'overall_recall': 0.6, 'overall_f1': 0.578219533275713, 'overall_accuracy': 0.8194253938832252}
			------------EPOCH 5---------------
Loss:  tensor(0.6972, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9103, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6185, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4401, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5577, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1415, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2136, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1830, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6000, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6565, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6239, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6107, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5056, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5029, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5442, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5207, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4178, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2103, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4250, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7414, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4795, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0167, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4245, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7925, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2241, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0428, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5369, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8396, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3882, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4391, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2851, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4241, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5462, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7169, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3643, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1922, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5893, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3141, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1509, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4999, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.47692307692307695, 'recall': 0.5714285714285714, 'f1': 0.519916142557652, 'number': 434}, 'P': {'precision': 0.6402266288951841, 'recall': 0.6637298091042585, 'f1': 0.6517664023071377, 'number': 681}, 'overall_precision': 0.5709624796084829, 'overall_recall': 0.6278026905829597, 'overall_f1': 0.598035027765912, 'overall_accuracy': 0.8207599629286376}
			------------EPOCH 6---------------
Loss:  tensor(0.5363, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7069, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4233, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2750, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4188, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0754, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1773, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1534, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5071, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5886, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4834, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5577, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3385, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4109, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3413, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2494, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1986, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6287, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3276, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8794, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3518, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6578, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1814, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0436, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4771, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7928, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2697, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3530, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2698, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3388, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4538, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6323, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3584, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1532, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6566, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3246, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1091, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4512, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.47368421052631576, 'recall': 0.5599078341013825, 'f1': 0.5131995776135163, 'number': 434}, 'P': {'precision': 0.6010854816824966, 'recall': 0.6505139500734214, 'f1': 0.6248236953455572, 'number': 681}, 'overall_precision': 0.5488, 'overall_recall': 0.6152466367713004, 'overall_f1': 0.5801268498942916, 'overall_accuracy': 0.8273957367933271}
			------------EPOCH 7---------------
Loss:  tensor(0.4130, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5526, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1895, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1226, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2226, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0319, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1609, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1035, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4124, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3830, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2948, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3892, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2365, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1919, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3025, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1714, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1731, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1445, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2818, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4367, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2818, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0371, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2309, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5233, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1363, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0265, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2261, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6603, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2339, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2719, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2649, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2402, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3826, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6573, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1951, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0704, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5157, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1773, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0630, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2469, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3900343642611684, 'recall': 0.5230414746543779, 'f1': 0.44685039370078744, 'number': 434}, 'P': {'precision': 0.5340086830680174, 'recall': 0.5418502202643172, 'f1': 0.5379008746355686, 'number': 681}, 'overall_precision': 0.46818538884524746, 'overall_recall': 0.5345291479820627, 'overall_f1': 0.4991624790619765, 'overall_accuracy': 0.7930676552363299}
			------------EPOCH 8---------------
Loss:  tensor(0.4624, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5818, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3411, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2260, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2904, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0743, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1173, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3541, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2475, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2141, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3986, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2254, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1593, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3552, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2209, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1172, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2329, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4111, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2303, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7483, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1965, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5221, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1888, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0177, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1892, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6065, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1954, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1642, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1540, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2039, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3806, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4972, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2865, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0931, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5086, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1985, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0644, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1722, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4873294346978557, 'recall': 0.576036866359447, 'f1': 0.5279831045406547, 'number': 434}, 'P': {'precision': 0.6347305389221557, 'recall': 0.6226138032305433, 'f1': 0.6286137879911046, 'number': 681}, 'overall_precision': 0.5707027942421676, 'overall_recall': 0.6044843049327354, 'overall_f1': 0.5871080139372822, 'overall_accuracy': 0.8147914735866543}
			------------EPOCH 9---------------
Loss:  tensor(0.2528, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5701, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1709, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0595, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1999, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0190, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1179, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0899, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2449, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1415, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3040, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1634, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0881, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2899, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2883, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1215, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0871, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3477, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1056, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6390, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1044, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4180, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1559, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0135, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2114, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7739, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3170, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3571, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1538, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3740, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3190, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4221, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1257, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0399, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4967, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1256, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1374, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4878, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5220588235294118, 'recall': 0.3271889400921659, 'f1': 0.4022662889518414, 'number': 434}, 'P': {'precision': 0.4969758064516129, 'recall': 0.723935389133627, 'f1': 0.5893604303646144, 'number': 681}, 'overall_precision': 0.502373417721519, 'overall_recall': 0.5695067264573991, 'overall_f1': 0.5338377469525011, 'overall_accuracy': 0.7962187210379982}
			------------EPOCH 10---------------
Loss:  tensor(0.5647, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6321, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1708, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1642, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0346, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0924, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0928, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2597, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2484, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2566, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3751, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3574, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2809, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2752, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2269, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0990, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1069, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2681, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1343, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6684, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1542, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4650, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1366, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0187, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0772, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5569, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0953, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1110, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1187, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1068, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2595, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3661, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2056, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0414, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4508, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0778, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0771, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2062, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5668316831683168, 'recall': 0.5276497695852534, 'f1': 0.5465393794749402, 'number': 434}, 'P': {'precision': 0.5960099750623441, 'recall': 0.7019089574155654, 'f1': 0.6446392447741065, 'number': 681}, 'overall_precision': 0.5862354892205639, 'overall_recall': 0.6340807174887892, 'overall_f1': 0.6092201637225334, 'overall_accuracy': 0.8265060240963855}
			------------EPOCH 11---------------
Loss:  tensor(0.1897, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4488, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1433, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0546, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1233, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0455, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0696, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0748, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2846, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1118, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1241, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2877, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1354, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0437, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1591, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0835, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0575, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0653, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1092, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1749, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0864, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5112, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0527, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3448, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1165, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0033, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0592, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4930, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1214, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0815, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0860, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0868, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2347, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3328, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0648, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0231, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3598, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0663, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0420, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0751, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5548245614035088, 'recall': 0.5829493087557603, 'f1': 0.5685393258426966, 'number': 434}, 'P': {'precision': 0.6084656084656085, 'recall': 0.6754772393538914, 'f1': 0.640222686151705, 'number': 681}, 'overall_precision': 0.5882838283828383, 'overall_recall': 0.6394618834080718, 'overall_f1': 0.6128061882251827, 'overall_accuracy': 0.8249490268767378}
			------------EPOCH 12---------------
Loss:  tensor(0.1036, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4093, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0511, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0176, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0445, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0098, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0752, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0760, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1383, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0617, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0921, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3026, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0723, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0522, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1807, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0597, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0474, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0666, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0674, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1678, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0390, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4649, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0560, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0994, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0178, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4209, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0330, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0450, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0723, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0462, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1935, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2964, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0521, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0205, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4572, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0824, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0330, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0468, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5386313465783664, 'recall': 0.5622119815668203, 'f1': 0.5501691093573844, 'number': 434}, 'P': {'precision': 0.6486486486486487, 'recall': 0.7048458149779736, 'f1': 0.6755805770584096, 'number': 681}, 'overall_precision': 0.6068734283319362, 'overall_recall': 0.6493273542600897, 'overall_f1': 0.6273830155979202, 'overall_accuracy': 0.8358850787766451}
			------------EPOCH 13---------------
Loss:  tensor(0.0917, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3560, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0424, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0134, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0263, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0031, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0558, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0559, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1531, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0288, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0528, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2258, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0552, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0138, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1367, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0361, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0419, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0524, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0966, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1511, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0324, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4440, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0311, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3114, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0941, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0127, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4098, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0209, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0248, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0705, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0246, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1423, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2241, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0207, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0098, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3128, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0459, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0403, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0413, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5496535796766744, 'recall': 0.5483870967741935, 'f1': 0.5490196078431372, 'number': 434}, 'P': {'precision': 0.6586741889985895, 'recall': 0.6857562408223201, 'f1': 0.6719424460431656, 'number': 681}, 'overall_precision': 0.617338003502627, 'overall_recall': 0.6322869955156951, 'overall_f1': 0.6247230837394772, 'overall_accuracy': 0.8328452270620945}
			------------EPOCH 14---------------
Loss:  tensor(0.0889, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3716, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0375, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0107, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0239, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0025, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0609, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0623, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1703, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0206, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0394, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2646, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0559, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0089, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1192, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0231, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0320, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0471, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0819, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1229, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0271, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4312, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0208, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2479, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0881, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0030, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0137, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3830, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0217, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0440, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0559, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0216, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1326, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2086, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0165, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0069, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3201, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0378, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0033, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0191, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5956284153005464, 'recall': 0.5023041474654378, 'f1': 0.5450000000000002, 'number': 434}, 'P': {'precision': 0.6666666666666666, 'recall': 0.6989720998531571, 'f1': 0.6824372759856631, 'number': 681}, 'overall_precision': 0.6425925925925926, 'overall_recall': 0.6224215246636772, 'overall_f1': 0.6323462414578587, 'overall_accuracy': 0.8288415199258573}
			------------EPOCH 15---------------
Loss:  tensor(0.0576, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3384, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0370, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0175, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0306, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0028, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0475, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0532, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1148, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0236, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0407, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2216, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0360, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0059, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0919, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0149, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0211, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0391, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0802, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0949, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0131, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4057, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0206, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2443, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0930, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0029, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0282, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4048, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0484, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0573, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0467, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0316, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1137, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1831, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0162, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2746, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0376, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0098, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5855614973262032, 'recall': 0.5046082949308756, 'f1': 0.5420792079207921, 'number': 434}, 'P': {'precision': 0.6813793103448276, 'recall': 0.7254038179148311, 'f1': 0.7027027027027026, 'number': 681}, 'overall_precision': 0.6487716105550501, 'overall_recall': 0.6394618834080718, 'overall_f1': 0.6440831074977416, 'overall_accuracy': 0.833660797034291}
			------------EPOCH 16---------------
Loss:  tensor(0.0345, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3291, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0277, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0062, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0286, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0029, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0484, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0577, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1939, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0315, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0395, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2722, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0343, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0092, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1031, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0143, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0189, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0302, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0431, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0777, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0086, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3577, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0151, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2072, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0833, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0029, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0241, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3838, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0498, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0737, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0613, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0503, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1518, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2122, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0298, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0101, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0608, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0106, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5904761904761905, 'recall': 0.5714285714285714, 'f1': 0.5807962529274004, 'number': 434}, 'P': {'precision': 0.6749654218533887, 'recall': 0.7165932452276065, 'f1': 0.6951566951566952, 'number': 681}, 'overall_precision': 0.6439195100612424, 'overall_recall': 0.6600896860986547, 'overall_f1': 0.6519043401240037, 'overall_accuracy': 0.8424467099165894}
			------------EPOCH 17---------------
Loss:  tensor(0.0288, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2861, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0156, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0026, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0186, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0452, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0468, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1184, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0161, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0477, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2483, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0780, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0335, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1302, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0397, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0211, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0369, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0590, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1138, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0071, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4089, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0124, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2256, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0785, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0083, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3345, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0347, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0401, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0445, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0418, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1416, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1801, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0492, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0192, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3229, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0497, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0044, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0296, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.49369747899159666, 'recall': 0.5414746543778802, 'f1': 0.5164835164835165, 'number': 434}, 'P': {'precision': 0.6138743455497382, 'recall': 0.6886930983847284, 'f1': 0.6491349480968859, 'number': 681}, 'overall_precision': 0.567741935483871, 'overall_recall': 0.6313901345291479, 'overall_f1': 0.5978768577494692, 'overall_accuracy': 0.8396663577386468}
			------------EPOCH 18---------------
Loss:  tensor(0.0398, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3404, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0213, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0053, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0162, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0446, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0413, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1372, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0096, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0186, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1876, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0181, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0057, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0806, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0258, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0287, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0382, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0904, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1050, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0459, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3667, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0468, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2180, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0714, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0040, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3216, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0105, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0240, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0365, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0104, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0838, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1429, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0146, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0078, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2852, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0369, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0562, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4789915966386555, 'recall': 0.5253456221198156, 'f1': 0.501098901098901, 'number': 434}, 'P': {'precision': 0.5376884422110553, 'recall': 0.6284875183553598, 'f1': 0.5795531482735274, 'number': 681}, 'overall_precision': 0.5157232704402516, 'overall_recall': 0.5883408071748879, 'overall_f1': 0.5496439044826142, 'overall_accuracy': 0.819870250231696}
			------------EPOCH 19---------------
Loss:  tensor(0.0717, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3032, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0373, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0203, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0220, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0411, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0387, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1254, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0253, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0260, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1711, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0269, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0057, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0569, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0103, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0098, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0253, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0358, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0714, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0503, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3090, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0187, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1789, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1063, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0271, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3292, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0249, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0515, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0392, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0258, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0791, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1473, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0126, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0033, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2209, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0340, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0097, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5646766169154229, 'recall': 0.5230414746543779, 'f1': 0.5430622009569378, 'number': 434}, 'P': {'precision': 0.6337135614702155, 'recall': 0.7342143906020558, 'f1': 0.6802721088435375, 'number': 681}, 'overall_precision': 0.6104114189756508, 'overall_recall': 0.6520179372197309, 'overall_f1': 0.6305290546400694, 'overall_accuracy': 0.8377386468952734}
			------------EPOCH 20---------------
Loss:  tensor(0.0223, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2371, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0100, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0071, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0389, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0376, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1113, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0116, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0125, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1783, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0252, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0061, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0536, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0108, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0168, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0204, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0566, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0495, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0055, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3394, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0098, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1612, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0654, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2706, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0091, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0221, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0217, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0064, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0688, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1193, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0079, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2221, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0697, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0082, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5721271393643031, 'recall': 0.5391705069124424, 'f1': 0.5551601423487545, 'number': 434}, 'P': {'precision': 0.6170212765957447, 'recall': 0.723935389133627, 'f1': 0.6662162162162162, 'number': 681}, 'overall_precision': 0.6018211920529801, 'overall_recall': 0.6520179372197309, 'overall_f1': 0.6259147653895825, 'overall_accuracy': 0.833141797961075}
			------------EPOCH 21---------------
Loss:  tensor(0.0199, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2737, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0120, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0042, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0368, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0338, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1215, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0059, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0093, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1646, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0195, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0028, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0516, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0089, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0075, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0196, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0340, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0392, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0035, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2805, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0087, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1347, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0578, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2549, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0069, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0190, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0252, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0050, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0641, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1101, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0067, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1810, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0329, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0034, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5932642487046632, 'recall': 0.5276497695852534, 'f1': 0.5585365853658537, 'number': 434}, 'P': {'precision': 0.6297709923664122, 'recall': 0.7268722466960352, 'f1': 0.6748466257668712, 'number': 681}, 'overall_precision': 0.6177474402730375, 'overall_recall': 0.6493273542600897, 'overall_f1': 0.6331438565806734, 'overall_accuracy': 0.8328822984244671}
			------------EPOCH 22---------------
Loss:  tensor(0.0156, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1996, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0068, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0033, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0482, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0377, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1062, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0065, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1448, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0180, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0541, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0081, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0059, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0186, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0507, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0501, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3378, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0074, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1557, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0511, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2305, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0054, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0194, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0173, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0085, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0645, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1000, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0061, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1952, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0325, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0049, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5828729281767956, 'recall': 0.4861751152073733, 'f1': 0.5301507537688442, 'number': 434}, 'P': {'precision': 0.6038882138517618, 'recall': 0.7298091042584435, 'f1': 0.660904255319149, 'number': 681}, 'overall_precision': 0.5974683544303797, 'overall_recall': 0.6349775784753363, 'overall_f1': 0.6156521739130435, 'overall_accuracy': 0.8255421686746988}
			------------EPOCH 23---------------
Loss:  tensor(0.0133, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2205, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0062, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0045, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0351, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0283, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1069, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0052, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0058, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1324, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0160, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0842, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0088, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0055, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0342, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0404, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0405, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2663, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0067, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1207, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0719, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3162, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0050, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0175, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0182, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0045, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0841, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1329, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0056, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1829, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0263, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0025, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.572972972972973, 'recall': 0.48847926267281105, 'f1': 0.527363184079602, 'number': 434}, 'P': {'precision': 0.5975903614457831, 'recall': 0.7283406754772394, 'f1': 0.6565188616810059, 'number': 681}, 'overall_precision': 0.59, 'overall_recall': 0.6349775784753363, 'overall_f1': 0.6116630669546437, 'overall_accuracy': 0.8295829471733086}
			------------EPOCH 24---------------
Loss:  tensor(0.0110, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2031, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0057, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0036, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0363, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0362, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0964, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0045, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0054, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1611, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0143, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0856, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0075, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0053, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0227, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0457, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0509, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2906, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0067, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1624, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0537, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2298, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0083, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0180, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0131, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0048, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0696, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1215, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0067, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2662, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0293, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0069, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5371549893842887, 'recall': 0.5829493087557603, 'f1': 0.5591160220994476, 'number': 434}, 'P': {'precision': 0.6490984743411928, 'recall': 0.6872246696035242, 'f1': 0.6676176890156919, 'number': 681}, 'overall_precision': 0.6048657718120806, 'overall_recall': 0.6466367713004484, 'overall_f1': 0.6250541829215431, 'overall_accuracy': 0.8381835032437442}
			------------EPOCH 25---------------
Loss:  tensor(0.0100, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3254, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0164, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0069, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0385, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0362, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0959, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0037, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0063, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1283, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0142, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0482, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0076, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0060, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0188, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0266, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0509, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2718, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0078, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1603, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0676, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2461, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0070, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0172, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0116, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0031, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0407, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0724, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0050, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1653, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0367, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0026, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5535714285714286, 'recall': 0.5714285714285714, 'f1': 0.562358276643991, 'number': 434}, 'P': {'precision': 0.655076495132128, 'recall': 0.6916299559471366, 'f1': 0.6728571428571428, 'number': 681}, 'overall_precision': 0.6161096829477292, 'overall_recall': 0.6448430493273543, 'overall_f1': 0.6301489921121823, 'overall_accuracy': 0.8359221501390176}
			------------EPOCH 26---------------
Loss:  tensor(0.0084, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2482, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0064, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0055, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0349, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0321, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0811, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0045, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1216, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0198, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0294, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0080, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0030, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0116, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0378, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0265, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2192, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0047, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1166, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0431, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2035, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0066, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0162, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0080, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0034, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0324, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0626, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0043, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1546, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0235, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0027, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5569007263922519, 'recall': 0.5299539170506913, 'f1': 0.5430932703659976, 'number': 434}, 'P': {'precision': 0.6333333333333333, 'recall': 0.7254038179148311, 'f1': 0.676249144421629, 'number': 681}, 'overall_precision': 0.6068734283319362, 'overall_recall': 0.6493273542600897, 'overall_f1': 0.6273830155979202, 'overall_accuracy': 0.8338832252085264}
			------------EPOCH 27---------------
Loss:  tensor(0.0057, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1608, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0067, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0035, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0192, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0181, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1111, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0059, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0814, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0167, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0205, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0093, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0108, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0397, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0188, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1795, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0041, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1057, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0414, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1654, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0064, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0164, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0033, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0036, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0196, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0499, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0039, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1317, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0257, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0014, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5912596401028277, 'recall': 0.5299539170506913, 'f1': 0.5589307411907656, 'number': 434}, 'P': {'precision': 0.6242038216560509, 'recall': 0.7195301027900147, 'f1': 0.6684856753069577, 'number': 681}, 'overall_precision': 0.6132879045996593, 'overall_recall': 0.6457399103139013, 'overall_f1': 0.6290956749672346, 'overall_accuracy': 0.8328822984244671}
			------------EPOCH 28---------------
Loss:  tensor(0.0041, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1405, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0031, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0199, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0180, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0876, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0837, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0143, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0141, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0088, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0064, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0447, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0162, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1691, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0037, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0801, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0272, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1393, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0062, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0144, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0036, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0225, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0484, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0036, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1630, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0276, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0013, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5634146341463414, 'recall': 0.532258064516129, 'f1': 0.5473933649289098, 'number': 434}, 'P': {'precision': 0.6354166666666666, 'recall': 0.7165932452276065, 'f1': 0.6735679779158039, 'number': 681}, 'overall_precision': 0.6103565365025467, 'overall_recall': 0.6448430493273543, 'overall_f1': 0.6271260357610118, 'overall_accuracy': 0.8335495829471733}
			------------EPOCH 29---------------
Loss:  tensor(0.0037, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1235, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0156, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0161, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0757, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0130, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0229, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0115, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0074, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0321, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0179, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1740, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0033, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0708, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0252, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1367, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0060, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0157, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0030, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0217, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0396, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0034, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1129, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0225, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0012, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5434298440979956, 'recall': 0.5622119815668203, 'f1': 0.5526613816534541, 'number': 434}, 'P': {'precision': 0.6498637602179836, 'recall': 0.7004405286343612, 'f1': 0.6742049469964664, 'number': 681}, 'overall_precision': 0.6094674556213018, 'overall_recall': 0.6466367713004484, 'overall_f1': 0.6275021758050479, 'overall_accuracy': 0.8365152919369787}
			------------EPOCH 30---------------
Loss:  tensor(0.0034, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1429, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0025, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0292, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0230, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0787, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0763, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0158, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0100, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0082, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0359, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0130, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1431, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0071, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0611, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0176, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1254, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0057, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0174, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0266, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0525, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0032, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1182, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0259, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0012, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5463182897862233, 'recall': 0.5299539170506913, 'f1': 0.5380116959064327, 'number': 434}, 'P': {'precision': 0.6261203585147247, 'recall': 0.7180616740088106, 'f1': 0.6689466484268125, 'number': 681}, 'overall_precision': 0.5981697171381032, 'overall_recall': 0.6448430493273543, 'overall_f1': 0.6206301251618471, 'overall_accuracy': 0.8321779425393884}
	Train size: 50 Test size: 50


		-------------RUN 1-----------
			------------EPOCH 1---------------
Loss:  tensor(3.2094, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.1375, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.9097, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4712, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.6206, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2267, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3709, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3455, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1157, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.6605, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.8014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.4970, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5843, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2746, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5351, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1722, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.3699, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2133, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2241, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.0509, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.1104, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.0495, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8640, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.0299, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.28611898016997167, 'recall': 0.08251633986928104, 'f1': 0.12809131261889664, 'number': 1224}, 'P': {'precision': 0.19924337957124844, 'recall': 0.1440291704649043, 'f1': 0.1671957671957672, 'number': 2194}, 'overall_precision': 0.2150593089221248, 'overall_recall': 0.12200117027501463, 'overall_f1': 0.1556841515773754, 'overall_accuracy': 0.7043495809492153}
			------------EPOCH 2---------------
Loss:  tensor(1.8246, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.9347, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.7704, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3125, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.7708, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1059, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2639, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2871, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7377, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.7965, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2054, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5482, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4869, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9391, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3542, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3896, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1224, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6033, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4328, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4475, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3913, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5718, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4489, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5977, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5474, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.40607734806629836, 'recall': 0.24019607843137256, 'f1': 0.30184804928131415, 'number': 1224}, 'P': {'precision': 0.49163498098859315, 'recall': 0.589334548769371, 'f1': 0.5360696517412935, 'number': 2194}, 'overall_precision': 0.4731663685152057, 'overall_recall': 0.46430661205383267, 'overall_f1': 0.46869462492616654, 'overall_accuracy': 0.7689878677081602}
			------------EPOCH 3---------------
Loss:  tensor(1.3389, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4835, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3235, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2406, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4157, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0697, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1781, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2635, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5434, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2980, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8574, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1417, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3541, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7281, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9322, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2498, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0841, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1169, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0136, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0084, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0057, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1206, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0222, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4213, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2125, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.36879432624113473, 'recall': 0.4673202614379085, 'f1': 0.41225225225225226, 'number': 1224}, 'P': {'precision': 0.6198267564966313, 'recall': 0.5870556061987238, 'f1': 0.602996254681648, 'number': 2194}, 'overall_precision': 0.512537889225682, 'overall_recall': 0.5441778818022235, 'overall_f1': 0.5278842060451255, 'overall_accuracy': 0.8021914100524704}
			------------EPOCH 4---------------
Loss:  tensor(0.9394, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1694, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0279, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0519, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1363, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2253, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3498, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9274, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6213, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7907, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2767, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4990, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7908, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2200, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0653, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7408, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6325, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5957, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6127, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9317, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7428, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3337, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8956, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3958041958041958, 'recall': 0.4624183006535948, 'f1': 0.426525998492841, 'number': 1224}, 'P': {'precision': 0.5940008449514153, 'recall': 0.6408386508659982, 'f1': 0.6165314623985968, 'number': 2194}, 'overall_precision': 0.519357387411114, 'overall_recall': 0.5769455822118198, 'overall_f1': 0.5466389466389467, 'overall_accuracy': 0.8120681877537453}
			------------EPOCH 5---------------
Loss:  tensor(0.6393, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7576, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6114, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1232, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0851, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0388, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0771, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2026, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2904, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5969, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3766, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5124, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2115, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3876, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4549, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1471, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0412, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5000, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3985, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3797, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3774, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7039, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4809, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1957, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6902, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.39777468706536856, 'recall': 0.4673202614379085, 'f1': 0.4297520661157025, 'number': 1224}, 'P': {'precision': 0.6002386634844868, 'recall': 0.6877848678213309, 'f1': 0.6410365335598981, 'number': 2194}, 'overall_precision': 0.5265688259109311, 'overall_recall': 0.6088355763604447, 'overall_f1': 0.5647218453188602, 'overall_accuracy': 0.8203661055580617}
			------------EPOCH 6---------------
Loss:  tensor(0.4320, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4691, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4055, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0829, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6192, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0220, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0525, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1300, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1817, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3639, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2422, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3676, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1569, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2164, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2728, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1166, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0162, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2812, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2178, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2362, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5790, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3157, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1416, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5740, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4389438943894389, 'recall': 0.434640522875817, 'f1': 0.4367816091954023, 'number': 1224}, 'P': {'precision': 0.5778364116094987, 'recall': 0.6987237921604376, 'f1': 0.6325562203424798, 'number': 2194}, 'overall_precision': 0.5342820181112549, 'overall_recall': 0.604154476301931, 'overall_f1': 0.5670740079637512, 'overall_accuracy': 0.8232033049217693}
			------------EPOCH 7---------------
Loss:  tensor(0.2614, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2841, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2312, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0458, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4522, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0093, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0257, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0946, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1371, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2281, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1450, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1676, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1406, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1559, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1394, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0097, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1758, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1418, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1891, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1446, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6113, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2605, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0965, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5046, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4904030710172745, 'recall': 0.41748366013071897, 'f1': 0.4510150044130627, 'number': 1224}, 'P': {'precision': 0.5755708590068865, 'recall': 0.723792160437557, 'f1': 0.6412275388653342, 'number': 2194}, 'overall_precision': 0.5522230991844251, 'overall_recall': 0.6141018139262727, 'overall_f1': 0.5815209862861892, 'overall_accuracy': 0.8212208267052874}
			------------EPOCH 8---------------
Loss:  tensor(0.1323, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1652, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1670, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0293, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4560, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0096, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0163, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1065, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1098, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2259, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1386, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1756, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1360, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1268, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1839, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0864, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0131, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1314, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1565, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0896, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4822, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2164, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0554, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4489, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3976311336717428, 'recall': 0.3839869281045752, 'f1': 0.39068994181213634, 'number': 1224}, 'P': {'precision': 0.5978004713275726, 'recall': 0.6937101185050136, 'f1': 0.6421940928270042, 'number': 2194}, 'overall_precision': 0.5343347639484979, 'overall_recall': 0.582796957284962, 'overall_f1': 0.5575146935348447, 'overall_accuracy': 0.8061444953583893}
			------------EPOCH 9---------------
Loss:  tensor(0.1690, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2302, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0992, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0315, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4739, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0075, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0091, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0637, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1055, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0946, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0757, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1296, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1414, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0865, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0880, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0131, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1621, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0924, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1307, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1088, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4710, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1975, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0309, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3490, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.46314831665150136, 'recall': 0.4158496732026144, 'f1': 0.438226431338786, 'number': 1224}, 'P': {'precision': 0.6025401568920433, 'recall': 0.735186873290793, 'f1': 0.662287004721823, 'number': 2194}, 'overall_precision': 0.5619703389830508, 'overall_recall': 0.6208308952603862, 'overall_f1': 0.589936057825966, 'overall_accuracy': 0.816994705477338}
			------------EPOCH 10---------------
Loss:  tensor(0.0527, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0665, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0693, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0145, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4258, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0044, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0157, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0806, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1092, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1501, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0747, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0845, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1305, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0840, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0443, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0757, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0042, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0535, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0949, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1281, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1265, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4162, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1792, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0221, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3275, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.49734513274336284, 'recall': 0.4591503267973856, 'f1': 0.4774851316907392, 'number': 1224}, 'P': {'precision': 0.6164114644680015, 'recall': 0.715587967183227, 'f1': 0.66230753005695, 'number': 2194}, 'overall_precision': 0.579820505847158, 'overall_recall': 0.6237565827969573, 'overall_f1': 0.6009866102889361, 'overall_accuracy': 0.8219330943279755}
			------------EPOCH 11---------------
Loss:  tensor(0.0463, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0732, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0552, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0107, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3240, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0025, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0253, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0678, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0897, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2538, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2425, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1471, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1153, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1165, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0935, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0845, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0055, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0855, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3046, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1744, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4364, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2146, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0460, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3055, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4752994011976048, 'recall': 0.5187908496732027, 'f1': 0.49609375000000006, 'number': 1224}, 'P': {'precision': 0.6242502142245073, 'recall': 0.6640838650865998, 'f1': 0.6435512367491166, 'number': 2194}, 'overall_precision': 0.5700272479564032, 'overall_recall': 0.6120538326506729, 'overall_f1': 0.5902934537246048, 'overall_accuracy': 0.8124718060732686}
			------------EPOCH 12---------------
Loss:  tensor(0.0894, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1308, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0108, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3510, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0049, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0154, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0605, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0652, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1385, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1307, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2186, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1217, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2074, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2800, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0909, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0121, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1465, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0693, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1409, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0946, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5310, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1846, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0310, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2970, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3900262467191601, 'recall': 0.6070261437908496, 'f1': 0.47491211249600507, 'number': 1224}, 'P': {'precision': 0.6633611691022965, 'recall': 0.5793072014585232, 'f1': 0.6184914841849148, 'number': 2194}, 'overall_precision': 0.5270871499607432, 'overall_recall': 0.5892334698654184, 'overall_f1': 0.5564304461942258, 'overall_accuracy': 0.7888601343811581}
			------------EPOCH 13---------------
Loss:  tensor(0.1535, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1134, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1711, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0303, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5376, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0180, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0153, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1148, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0945, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1835, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1331, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1420, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1065, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0916, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0108, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0544, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0638, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0659, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1115, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4658, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2774, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0575, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3091, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.41368078175895767, 'recall': 0.5187908496732027, 'f1': 0.4603117071402682, 'number': 1224}, 'P': {'precision': 0.5462735462735463, 'recall': 0.6080218778486782, 'f1': 0.5754961173425367, 'number': 2194}, 'overall_precision': 0.49509680663816946, 'overall_recall': 0.5760678759508484, 'overall_f1': 0.5325219743069642, 'overall_accuracy': 0.8129585222821054}
			------------EPOCH 14---------------
Loss:  tensor(0.1066, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1558, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1299, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0096, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3342, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0037, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0052, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0561, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0552, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0469, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0462, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0406, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1105, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1099, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0810, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0900, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0039, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0999, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0839, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1524, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1094, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3879, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1252, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0118, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1897, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.44818304172274565, 'recall': 0.5441176470588235, 'f1': 0.49151291512915135, 'number': 1224}, 'P': {'precision': 0.6605584642233857, 'recall': 0.6900638103919782, 'f1': 0.6749888542131076, 'number': 2194}, 'overall_precision': 0.5770248808893594, 'overall_recall': 0.6377998829724986, 'overall_f1': 0.6058921623123958, 'overall_accuracy': 0.8272038747358674}
			------------EPOCH 15---------------
Loss:  tensor(0.0187, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0300, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0292, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0034, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0057, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0608, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0614, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0765, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0615, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0671, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0966, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0610, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0313, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0624, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0284, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0142, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0275, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0152, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3110, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1062, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0055, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1758, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.48306997742663654, 'recall': 0.5245098039215687, 'f1': 0.5029377203290246, 'number': 1224}, 'P': {'precision': 0.6372629843363561, 'recall': 0.7046490428441203, 'f1': 0.6692640692640693, 'number': 2194}, 'overall_precision': 0.5826897470039947, 'overall_recall': 0.6401404330017554, 'overall_f1': 0.6100655234908685, 'overall_accuracy': 0.8290795128089461}
			------------EPOCH 16---------------
Loss:  tensor(0.0115, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0202, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0098, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2522, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0476, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0375, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0196, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0212, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0107, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0836, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0483, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0155, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0540, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0176, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0071, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0225, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0091, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2992, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0033, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1511, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.49135987978963186, 'recall': 0.5343137254901961, 'f1': 0.5119373776908024, 'number': 1224}, 'P': {'precision': 0.6543993231810491, 'recall': 0.7051048313582498, 'f1': 0.6788064940763492, 'number': 2194}, 'overall_precision': 0.5956698240866035, 'overall_recall': 0.6439438267992978, 'overall_f1': 0.6188668634893856, 'overall_accuracy': 0.8278805289774211}
			------------EPOCH 17---------------
Loss:  tensor(0.0082, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0140, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0064, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2258, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0442, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0312, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0159, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0180, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0073, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0814, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0395, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0094, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0509, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0134, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0047, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0139, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0053, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2741, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0820, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0025, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1318, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4696645253390435, 'recall': 0.5375816993464052, 'f1': 0.5013333333333333, 'number': 1224}, 'P': {'precision': 0.6464941569282137, 'recall': 0.7060164083865087, 'f1': 0.6749455337690632, 'number': 2194}, 'overall_precision': 0.5812483539636555, 'overall_recall': 0.6456992393212405, 'overall_f1': 0.6117810117810119, 'overall_accuracy': 0.8277974310881075}
			------------EPOCH 18---------------
Loss:  tensor(0.0071, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0113, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0053, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2126, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0441, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0278, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0134, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0154, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0054, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0736, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0327, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0071, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0457, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0112, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0107, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0039, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2513, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0752, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1189, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.48677248677248675, 'recall': 0.5261437908496732, 'f1': 0.5056929721240674, 'number': 1224}, 'P': {'precision': 0.6458852867830424, 'recall': 0.7082953509571559, 'f1': 0.6756521739130434, 'number': 2194}, 'overall_precision': 0.5894341646554035, 'overall_recall': 0.6430661205383265, 'overall_f1': 0.6150832517140059, 'overall_accuracy': 0.82824853391581}
			------------EPOCH 19---------------
Loss:  tensor(0.0053, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0097, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0041, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2045, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0414, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0225, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0111, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0127, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0039, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0671, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0291, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0060, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0415, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0105, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0031, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0077, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0032, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2336, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0674, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1082, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.48502994011976047, 'recall': 0.5294117647058824, 'f1': 0.50625, 'number': 1224}, 'P': {'precision': 0.6436258278145696, 'recall': 0.7087511394712853, 'f1': 0.6746203904555315, 'number': 2194}, 'overall_precision': 0.5871535181236673, 'overall_recall': 0.6445289643066121, 'overall_f1': 0.6145048814504881, 'overall_accuracy': 0.8298511360668582}
			------------EPOCH 20---------------
Loss:  tensor(0.0040, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0085, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0034, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1870, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0436, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0249, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0100, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0104, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0029, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0718, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0236, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0496, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0100, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0063, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2179, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0672, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1071, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4559023066485753, 'recall': 0.5490196078431373, 'f1': 0.4981467753891772, 'number': 1224}, 'P': {'precision': 0.6562097120756338, 'recall': 0.6959890610756609, 'f1': 0.6755142667551427, 'number': 2194}, 'overall_precision': 0.5785319652722968, 'overall_recall': 0.6433586892919836, 'overall_f1': 0.6092256545227872, 'overall_accuracy': 0.8262660556993281}
			------------EPOCH 21---------------
Loss:  tensor(0.0029, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0077, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0031, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2215, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0474, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0202, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0091, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0103, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0025, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0595, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0271, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0044, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0376, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0100, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0028, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0062, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3550, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0970, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1133, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4802052785923754, 'recall': 0.5351307189542484, 'f1': 0.5061823802163833, 'number': 1224}, 'P': {'precision': 0.6429458005792305, 'recall': 0.7082953509571559, 'f1': 0.6740403383214053, 'number': 2194}, 'overall_precision': 0.5842369743454112, 'overall_recall': 0.6462843768285547, 'overall_f1': 0.6136963467148214, 'overall_accuracy': 0.827465039530853}
			------------EPOCH 22---------------
Loss:  tensor(0.0029, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0071, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0031, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1911, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0469, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0291, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0097, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0079, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0035, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1271, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0260, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0062, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0713, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0097, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0032, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0067, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0033, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3309, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0741, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1153, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4550938337801609, 'recall': 0.5547385620915033, 'f1': 0.5, 'number': 1224}, 'P': {'precision': 0.6565481352992194, 'recall': 0.6900638103919782, 'f1': 0.6728888888888889, 'number': 2194}, 'overall_precision': 0.5774091627172195, 'overall_recall': 0.6416032767700409, 'overall_f1': 0.6078159645232815, 'overall_accuracy': 0.8253638500439232}
			------------EPOCH 23---------------
Loss:  tensor(0.0034, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0074, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0033, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1947, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0436, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0249, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0082, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0080, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0634, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0249, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0040, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0412, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0099, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0025, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0061, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0026, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2169, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0602, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0969, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4964871194379391, 'recall': 0.5196078431372549, 'f1': 0.5077844311377246, 'number': 1224}, 'P': {'precision': 0.6310679611650486, 'recall': 0.7110300820419325, 'f1': 0.6686669524217745, 'number': 2194}, 'overall_precision': 0.5851318944844125, 'overall_recall': 0.6424809830310123, 'overall_f1': 0.612466880490866, 'overall_accuracy': 0.8299104917020822}
			------------EPOCH 24---------------
Loss:  tensor(0.0024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0065, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0025, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1623, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0383, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0138, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0073, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0055, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0514, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0204, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0331, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0086, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0055, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2102, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0531, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0920, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.45734265734265733, 'recall': 0.5343137254901961, 'f1': 0.49284099472494347, 'number': 1224}, 'P': {'precision': 0.6447647308181432, 'recall': 0.6932543299908842, 'f1': 0.6681309027015154, 'number': 2194}, 'overall_precision': 0.5740300870942201, 'overall_recall': 0.636337039204213, 'overall_f1': 0.6035798529207714, 'overall_accuracy': 0.8265509627484033}
			------------EPOCH 25---------------
Loss:  tensor(0.0019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0056, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1338, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0365, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0133, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0069, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0044, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0546, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0193, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0292, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0082, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1856, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0497, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0874, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4871601208459215, 'recall': 0.5269607843137255, 'f1': 0.5062794348508635, 'number': 1224}, 'P': {'precision': 0.640728476821192, 'recall': 0.7055606198723792, 'f1': 0.6715835140997831, 'number': 2194}, 'overall_precision': 0.5863636363636363, 'overall_recall': 0.6416032767700409, 'overall_f1': 0.6127409891031014, 'overall_accuracy': 0.8289489304114532}
			------------EPOCH 26---------------
Loss:  tensor(0.0017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1172, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0354, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0163, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0068, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0033, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0515, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0151, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0031, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0257, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0069, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0048, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1819, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0424, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0688, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4752906976744186, 'recall': 0.5343137254901961, 'f1': 0.5030769230769231, 'number': 1224}, 'P': {'precision': 0.6380316930775647, 'recall': 0.6973564266180492, 'f1': 0.6663763066202091, 'number': 2194}, 'overall_precision': 0.5786963434022258, 'overall_recall': 0.638970157987127, 'overall_f1': 0.60734149054505, 'overall_accuracy': 0.8264678648590897}
			------------EPOCH 27---------------
Loss:  tensor(0.0015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0046, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1080, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0329, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0146, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0065, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0028, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0410, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0112, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0044, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0243, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0063, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0046, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1634, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0417, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0604, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4624113475177305, 'recall': 0.5326797385620915, 'f1': 0.49506454062262717, 'number': 1224}, 'P': {'precision': 0.6358718268830629, 'recall': 0.6964448495897904, 'f1': 0.6647813791603219, 'number': 2194}, 'overall_precision': 0.5717282979281406, 'overall_recall': 0.6377998829724986, 'overall_f1': 0.6029594800165953, 'overall_accuracy': 0.8275125240390323}
			------------EPOCH 28---------------
Loss:  tensor(0.0014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0041, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0963, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0307, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0105, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0063, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0462, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0112, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0044, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0253, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0054, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0042, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1464, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0282, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0493, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4753265602322206, 'recall': 0.5351307189542484, 'f1': 0.5034588777863181, 'number': 1224}, 'P': {'precision': 0.6439106616097766, 'recall': 0.6964448495897904, 'f1': 0.6691482373549377, 'number': 2194}, 'overall_precision': 0.5819781391628899, 'overall_recall': 0.6386775892334698, 'overall_f1': 0.6090110196680151, 'overall_accuracy': 0.8265509627484033}
			------------EPOCH 29---------------
Loss:  tensor(0.0013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0890, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0311, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0071, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0062, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0405, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0104, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0041, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0219, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0053, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0039, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1399, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0283, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0388, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.46164874551971324, 'recall': 0.5261437908496732, 'f1': 0.49179075983199694, 'number': 1224}, 'P': {'precision': 0.6303453947368421, 'recall': 0.6987237921604376, 'f1': 0.6627756160830092, 'number': 2194}, 'overall_precision': 0.5688528873791482, 'overall_recall': 0.6369221767115272, 'overall_f1': 0.6009661835748792, 'overall_accuracy': 0.8260998599207009}
			------------EPOCH 30---------------
Loss:  tensor(0.0012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0033, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0904, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0293, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0075, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0060, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0485, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0078, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0039, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0335, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0050, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0036, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1423, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0353, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0487, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4950343773873186, 'recall': 0.5294117647058824, 'f1': 0.5116462692459534, 'number': 1224}, 'P': {'precision': 0.6361771944216571, 'recall': 0.7069279854147675, 'f1': 0.6696891191709844, 'number': 2194}, 'overall_precision': 0.5868694955964772, 'overall_recall': 0.6433586892919836, 'overall_f1': 0.6138171667829727, 'overall_accuracy': 0.8292338374605285}


		-------------RUN 2-----------
			------------EPOCH 1---------------
Loss:  tensor(3.2353, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.2911, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.0012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7450, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.9567, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.7648, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2632, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.7209, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0685, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.6959, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.3903, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.6133, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.7460, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3135, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.4379, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.3646, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2845, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2737, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2213, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9746, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.1965, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.0339, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.0273, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.1476, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.8564, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.39325842696629215, 'recall': 0.0348605577689243, 'f1': 0.06404391582799633, 'number': 1004}, 'P': {'precision': 0.3787751677852349, 'recall': 0.5283791691047396, 'f1': 0.44124114341558757, 'number': 1709}, 'overall_precision': 0.37929640113222807, 'overall_recall': 0.3457427202359012, 'overall_f1': 0.3617431546471268, 'overall_accuracy': 0.7029640718562874}
			------------EPOCH 2---------------
Loss:  tensor(1.8299, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.8577, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.7374, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4706, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.7455, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.7601, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5345, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6418, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7300, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6743, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6545, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.7436, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2316, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3178, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.7582, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5997, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5801, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.8318, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5994, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6880, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6611, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4491, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4853, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6881, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4846, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3238512035010941, 'recall': 0.14741035856573706, 'f1': 0.20260095824777552, 'number': 1004}, 'P': {'precision': 0.49239713774597493, 'recall': 0.644236395552955, 'f1': 0.5581749049429657, 'number': 1709}, 'overall_precision': 0.4637950241366506, 'overall_recall': 0.46037596756358273, 'overall_f1': 0.462079171291158, 'overall_accuracy': 0.7358832335329342}
			------------EPOCH 3---------------
Loss:  tensor(1.3528, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3717, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2981, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3890, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3194, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3812, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1975, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1959, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5628, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2735, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3760, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3762, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9958, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2968, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4066, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2253, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1847, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3794, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1964, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4700, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2835, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1225, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2831, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1972, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.37877312560856863, 'recall': 0.38745019920318724, 'f1': 0.3830625307730183, 'number': 1004}, 'P': {'precision': 0.541134389371487, 'recall': 0.6196606202457577, 'f1': 0.5777414075286416, 'number': 1709}, 'overall_precision': 0.48525469168900803, 'overall_recall': 0.5337265020272761, 'overall_f1': 0.5083377216078638, 'overall_accuracy': 0.788248502994012}
			------------EPOCH 4---------------
Loss:  tensor(0.9920, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0266, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0119, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3178, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0303, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1606, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9168, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8365, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4312, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9772, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0870, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0349, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7233, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2762, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1424, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9366, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8945, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0593, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8411, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3312, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0769, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8847, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7435, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8752, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8813, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.387719298245614, 'recall': 0.4402390438247012, 'f1': 0.41231343283582095, 'number': 1004}, 'P': {'precision': 0.580110497237569, 'recall': 0.6143943826799297, 'f1': 0.5967604433077579, 'number': 1709}, 'overall_precision': 0.5057627118644068, 'overall_recall': 0.5499447106524143, 'overall_f1': 0.5269291894755431, 'overall_accuracy': 0.7957934131736527}
			------------EPOCH 5---------------
Loss:  tensor(0.7048, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8671, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7400, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2868, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7970, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9548, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5913, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3663, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7346, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8339, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7525, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5198, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2611, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8736, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7555, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6888, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8607, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5571, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2476, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8832, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6353, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5185, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6365, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7955, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.45684394071490847, 'recall': 0.5219123505976095, 'f1': 0.48721524872152483, 'number': 1004}, 'P': {'precision': 0.5951871657754011, 'recall': 0.6512580456407255, 'f1': 0.6219614417435038, 'number': 1709}, 'overall_precision': 0.5425919787868744, 'overall_recall': 0.6033910799852562, 'overall_f1': 0.5713787085514834, 'overall_accuracy': 0.8121107784431137}
			------------EPOCH 6---------------
Loss:  tensor(0.4989, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5759, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5691, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2390, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5593, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9142, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4777, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4062, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2411, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4723, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7083, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5988, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3898, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2342, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7905, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5791, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5437, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7575, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4073, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1755, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7247, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4227, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3929, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4687, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6792, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3890295358649789, 'recall': 0.4591633466135458, 'f1': 0.4211968935587026, 'number': 1004}, 'P': {'precision': 0.5016151361329027, 'recall': 0.6360444704505559, 'f1': 0.5608875128998968, 'number': 1709}, 'overall_precision': 0.4618138424821002, 'overall_recall': 0.5705860670844084, 'overall_f1': 0.510469909315746, 'overall_accuracy': 0.8102544910179641}
			------------EPOCH 7---------------
Loss:  tensor(0.3174, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3566, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3538, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1820, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3929, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7105, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3686, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3490, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1826, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6427, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6524, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3909, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2030, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6226, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4462, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3820, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5607, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5952, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2434, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7745, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6719, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2953, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2999, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7020, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5011185682326622, 'recall': 0.22310756972111553, 'f1': 0.3087525844245348, 'number': 1004}, 'P': {'precision': 0.4822104466313399, 'recall': 0.7454651843183148, 'f1': 0.5856125028729028, 'number': 1709}, 'overall_precision': 0.4849465846552282, 'overall_recall': 0.5521562845558422, 'overall_f1': 0.5163736642537056, 'overall_accuracy': 0.770688622754491}
			------------EPOCH 8---------------
Loss:  tensor(0.8250, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8449, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7177, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3179, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5070, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7608, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3198, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3336, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1956, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4182, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8764, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4973, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3691, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1651, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5397, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5939, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4058, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6598, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3753, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1812, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6841, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3833, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3089, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3605, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7334, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4713839750260146, 'recall': 0.45119521912350596, 'f1': 0.4610687022900764, 'number': 1004}, 'P': {'precision': 0.6191222570532915, 'recall': 0.6933879461673493, 'f1': 0.6541540160088325, 'number': 1709}, 'overall_precision': 0.5697391304347826, 'overall_recall': 0.6037596756358274, 'overall_f1': 0.5862562634216177, 'overall_accuracy': 0.8025299401197605}
			------------EPOCH 9---------------
Loss:  tensor(0.2545, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3214, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2420, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1910, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2383, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5838, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2028, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2118, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0985, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3877, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2311, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1927, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1166, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3438, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2660, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1874, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3845, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1583, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0896, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4492, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1970, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2628, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1983, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6012, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.42970611596505165, 'recall': 0.5388446215139442, 'f1': 0.478126380910296, 'number': 1004}, 'P': {'precision': 0.6082974137931034, 'recall': 0.660620245757753, 'f1': 0.6333800841514726, 'number': 1709}, 'overall_precision': 0.536115569823435, 'overall_recall': 0.6155547364541099, 'overall_f1': 0.573095401509952, 'overall_accuracy': 0.796182634730539}
			------------EPOCH 10---------------
Loss:  tensor(0.2067, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2199, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1611, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1408, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1025, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4854, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1307, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0664, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1450, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2551, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1537, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0797, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2735, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1973, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2797, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0956, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0862, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3472, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1113, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1179, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0922, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4755, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4002770083102493, 'recall': 0.5756972111553785, 'f1': 0.47222222222222227, 'number': 1004}, 'P': {'precision': 0.616234124792932, 'recall': 0.6530134581626682, 'f1': 0.634090909090909, 'number': 1709}, 'overall_precision': 0.5204301075268817, 'overall_recall': 0.6244010320678216, 'overall_f1': 0.5676943699731903, 'overall_accuracy': 0.783188622754491}
			------------EPOCH 11---------------
Loss:  tensor(0.1842, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2000, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3282, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1532, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1710, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4703, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1135, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1075, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0311, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0710, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1953, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0823, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1164, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0684, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3125, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1242, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3851, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3976, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3725, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3250, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1332, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1048, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0777, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4349, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.464804469273743, 'recall': 0.41434262948207173, 'f1': 0.4381253291205898, 'number': 1004}, 'P': {'precision': 0.5839888940305414, 'recall': 0.7384435342305442, 'f1': 0.6521963824289405, 'number': 1709}, 'overall_precision': 0.5490837696335078, 'overall_recall': 0.6185035016586804, 'overall_f1': 0.5817299358641012, 'overall_accuracy': 0.8044760479041916}
			------------EPOCH 12---------------
Loss:  tensor(0.0686, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1288, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0739, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1279, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1042, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4451, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1758, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1791, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0524, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2333, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3362, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1933, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1461, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0879, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3174, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1117, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2428, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0945, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0712, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2797, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0980, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1175, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1329, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4748, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.526829268292683, 'recall': 0.4302788844621514, 'f1': 0.47368421052631576, 'number': 1004}, 'P': {'precision': 0.6012974051896207, 'recall': 0.7050906963136337, 'f1': 0.64907083221115, 'number': 1709}, 'overall_precision': 0.5796742209631728, 'overall_recall': 0.6033910799852562, 'overall_f1': 0.591294925049666, 'overall_accuracy': 0.8058832335329341}
			------------EPOCH 13---------------
Loss:  tensor(0.1327, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1325, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0825, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1205, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0494, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4437, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0843, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0591, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0205, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0806, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1899, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0634, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0838, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0625, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2522, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0664, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0607, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2112, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0535, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0635, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2803, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0355, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0610, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0361, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4235, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.42075623491552694, 'recall': 0.5209163346613546, 'f1': 0.4655095683133066, 'number': 1004}, 'P': {'precision': 0.5859576664945793, 'recall': 0.6641310708016384, 'f1': 0.6226001097092704, 'number': 1709}, 'overall_precision': 0.5213836477987421, 'overall_recall': 0.6111315886472539, 'overall_f1': 0.5627015102664178, 'overall_accuracy': 0.8163173652694611}
			------------EPOCH 14---------------
Loss:  tensor(0.0379, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0971, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0541, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1241, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0544, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3889, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0590, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0467, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0147, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0437, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1490, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0372, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0444, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0559, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1686, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0332, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0225, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1718, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0466, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0557, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2203, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0266, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0430, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0274, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4007, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5383734249713631, 'recall': 0.4681274900398406, 'f1': 0.500799147575919, 'number': 1004}, 'P': {'precision': 0.610337972166998, 'recall': 0.7185488589818607, 'f1': 0.6600376242945444, 'number': 1709}, 'overall_precision': 0.5885615251299827, 'overall_recall': 0.6258754146701069, 'overall_f1': 0.6066452304394426, 'overall_accuracy': 0.8140419161676646}
			------------EPOCH 15---------------
Loss:  tensor(0.0354, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0879, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0358, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1086, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0207, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3326, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0164, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0172, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0058, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0114, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1224, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0222, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0470, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0592, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1280, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0273, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0176, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1636, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0278, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0519, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2031, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0128, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0361, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0118, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3705, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.47080630213160335, 'recall': 0.5059760956175299, 'f1': 0.4877580412866058, 'number': 1004}, 'P': {'precision': 0.5936715218483174, 'recall': 0.6916325336454067, 'f1': 0.6389189189189189, 'number': 1709}, 'overall_precision': 0.5504885993485342, 'overall_recall': 0.6229266494655363, 'overall_f1': 0.5844717274770881, 'overall_accuracy': 0.8158682634730539}
			------------EPOCH 16---------------
Loss:  tensor(0.0289, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0664, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0164, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1034, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0100, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3118, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0113, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0139, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0046, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0087, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1080, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0170, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0690, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0472, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1194, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0174, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0103, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1496, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0221, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0457, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1836, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0082, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0233, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0080, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3564, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4980879541108987, 'recall': 0.5189243027888446, 'f1': 0.5082926829268293, 'number': 1004}, 'P': {'precision': 0.6173549398849973, 'recall': 0.6910473961380924, 'f1': 0.6521258972943125, 'number': 1709}, 'overall_precision': 0.5751943224062184, 'overall_recall': 0.6273497972723922, 'overall_f1': 0.6001410437235543, 'overall_accuracy': 0.8149850299401198}
			------------EPOCH 17---------------
Loss:  tensor(0.0135, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0584, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0129, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0999, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0066, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2959, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0088, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0118, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0042, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0070, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1054, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0286, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0707, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0491, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1670, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0143, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0064, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1373, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0316, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0428, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1627, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0061, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0179, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0064, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3387, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5142857142857142, 'recall': 0.48406374501992033, 'f1': 0.49871729091841965, 'number': 1004}, 'P': {'precision': 0.5994137762579385, 'recall': 0.7179637214745466, 'f1': 0.6533546325878595, 'number': 1709}, 'overall_precision': 0.5725267379679144, 'overall_recall': 0.6314043494286767, 'overall_f1': 0.6005258545135845, 'overall_accuracy': 0.8162275449101797}
			------------EPOCH 18---------------
Loss:  tensor(0.0115, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0451, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0117, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0061, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2837, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0064, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0104, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0047, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0058, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0955, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0192, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0498, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0437, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1472, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0142, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0068, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1331, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0136, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0413, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1568, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0055, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0271, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0054, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3322, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4817813765182186, 'recall': 0.47410358565737054, 'f1': 0.4779116465863454, 'number': 1004}, 'P': {'precision': 0.5891548607718613, 'recall': 0.7056758338209479, 'f1': 0.6421725239616614, 'number': 1709}, 'overall_precision': 0.5542009884678748, 'overall_recall': 0.6199778842609657, 'overall_f1': 0.5852470424495477, 'overall_accuracy': 0.815688622754491}
			------------EPOCH 19---------------
Loss:  tensor(0.0107, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0358, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0104, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0934, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0050, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2711, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0060, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0092, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0026, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0049, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0812, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0176, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0342, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0449, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0973, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0103, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0044, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1171, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0291, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0370, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1395, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0045, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0128, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0047, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3270, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5, 'recall': 0.5029880478087649, 'f1': 0.5014895729890765, 'number': 1004}, 'P': {'precision': 0.6125385405960946, 'recall': 0.6974839087185488, 'f1': 0.6522571819425446, 'number': 1709}, 'overall_precision': 0.5740866035182679, 'overall_recall': 0.6255068190195355, 'overall_f1': 0.5986946551420003, 'overall_accuracy': 0.8150299401197605}
			------------EPOCH 20---------------
Loss:  tensor(0.0096, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0313, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0089, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0910, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2584, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0046, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0080, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0039, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0778, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0161, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0282, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0461, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0862, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0088, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1189, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0144, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0398, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1392, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0214, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0040, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3032, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.49904397705544934, 'recall': 0.5199203187250996, 'f1': 0.5092682926829268, 'number': 1004}, 'P': {'precision': 0.6127017178552837, 'recall': 0.6887068461088356, 'f1': 0.6484848484848484, 'number': 1709}, 'overall_precision': 0.5726322885069093, 'overall_recall': 0.6262440103206782, 'overall_f1': 0.5982394366197183, 'overall_accuracy': 0.8161676646706587}
			------------EPOCH 21---------------
Loss:  tensor(0.0088, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0263, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0083, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0869, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0033, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2669, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0040, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0074, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0034, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0813, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0146, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0306, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0452, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0858, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0068, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0032, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1030, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0155, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0376, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1331, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0033, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0183, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0035, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3245, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5306818181818181, 'recall': 0.4651394422310757, 'f1': 0.4957537154989384, 'number': 1004}, 'P': {'precision': 0.5664467483506126, 'recall': 0.7033352837916911, 'f1': 0.6275123988514748, 'number': 1709}, 'overall_precision': 0.555962691538974, 'overall_recall': 0.6151861408035385, 'overall_f1': 0.584076990376203, 'overall_accuracy': 0.8104041916167665}
			------------EPOCH 22---------------
Loss:  tensor(0.0083, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0233, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0073, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0984, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0030, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2813, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0036, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0065, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0031, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0753, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0111, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0305, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0585, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0855, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0047, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0028, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1953, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0270, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0583, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1918, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0033, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0133, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0034, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3902, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.46660567246111617, 'recall': 0.5079681274900398, 'f1': 0.4864091559370529, 'number': 1004}, 'P': {'precision': 0.6044815007816571, 'recall': 0.6787595084844938, 'f1': 0.6394707828004411, 'number': 1709}, 'overall_precision': 0.5544488711819389, 'overall_recall': 0.6155547364541099, 'overall_f1': 0.5834061135371179, 'overall_accuracy': 0.8133083832335329}
			------------EPOCH 23---------------
Loss:  tensor(0.0094, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0209, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0099, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0899, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0037, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2836, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0039, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0068, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0025, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1167, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0088, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0291, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0694, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0852, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0086, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0037, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1627, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0134, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0581, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1598, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0031, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0257, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0039, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2927, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5238095238095238, 'recall': 0.4820717131474104, 'f1': 0.5020746887966806, 'number': 1004}, 'P': {'precision': 0.5830491962981004, 'recall': 0.70040959625512, 'f1': 0.6363636363636365, 'number': 1709}, 'overall_precision': 0.5646624118239839, 'overall_recall': 0.6196092886103944, 'overall_f1': 0.5908611599297012, 'overall_accuracy': 0.809251497005988}
			------------EPOCH 24---------------
Loss:  tensor(0.0090, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0212, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0080, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0798, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0026, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2332, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0028, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0055, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0684, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0071, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0351, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0408, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1056, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0040, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0941, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0193, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0368, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1258, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0140, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0028, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3014, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4714285714285714, 'recall': 0.49302788844621515, 'f1': 0.4819863680623174, 'number': 1004}, 'P': {'precision': 0.610683102208526, 'recall': 0.6957284961966061, 'f1': 0.6504376367614879, 'number': 1709}, 'overall_precision': 0.5618952285618952, 'overall_recall': 0.6207150755621084, 'overall_f1': 0.5898423817863396, 'overall_accuracy': 0.8159431137724551}
			------------EPOCH 25---------------
Loss:  tensor(0.0074, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0160, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0062, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0764, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0025, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2197, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0045, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0025, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0592, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0070, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0500, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0489, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1423, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0041, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0939, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0145, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0370, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1120, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0149, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0026, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2949, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.46649029982363316, 'recall': 0.5268924302788844, 'f1': 0.4948550046772685, 'number': 1004}, 'P': {'precision': 0.598421052631579, 'recall': 0.6653013458162669, 'f1': 0.6300914380714879, 'number': 1709}, 'overall_precision': 0.5491100856954515, 'overall_recall': 0.6140803538518246, 'overall_f1': 0.5797807551766139, 'overall_accuracy': 0.8125149700598803}
			------------EPOCH 26---------------
Loss:  tensor(0.0078, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0131, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0068, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0717, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0028, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2139, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0034, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0059, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0035, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0625, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0071, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0487, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0424, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1435, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0042, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0826, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0192, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0367, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1183, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0149, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0025, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2559, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5216515609264853, 'recall': 0.5159362549800797, 'f1': 0.5187781672508762, 'number': 1004}, 'P': {'precision': 0.604261796042618, 'recall': 0.6968987712112347, 'f1': 0.6472826086956522, 'number': 1709}, 'overall_precision': 0.5765856950067476, 'overall_recall': 0.6299299668263915, 'overall_f1': 0.6020785626211027, 'overall_accuracy': 0.8157035928143712}
			------------EPOCH 27---------------
Loss:  tensor(0.0071, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0118, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0059, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0717, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0048, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0681, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0062, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0288, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0390, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0936, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0031, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0705, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0124, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0276, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1034, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0195, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0025, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2406, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.48704663212435234, 'recall': 0.4681274900398406, 'f1': 0.4773996952767902, 'number': 1004}, 'P': {'precision': 0.5822906641000962, 'recall': 0.7080163838502048, 'f1': 0.6390282545550569, 'number': 1709}, 'overall_precision': 0.5520867564903056, 'overall_recall': 0.6192406929598231, 'overall_f1': 0.5837387074357192, 'overall_accuracy': 0.815748502994012}
			------------EPOCH 28---------------
Loss:  tensor(0.0066, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0100, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0052, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0662, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1861, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0059, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0388, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0054, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0317, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0398, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0959, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0612, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0181, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0208, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0788, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0123, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2336, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5030181086519114, 'recall': 0.49800796812749004, 'f1': 0.5005005005005004, 'number': 1004}, 'P': {'precision': 0.6120469627360898, 'recall': 0.7015798712697484, 'f1': 0.6537622682660851, 'number': 1709}, 'overall_precision': 0.5753471046393498, 'overall_recall': 0.6262440103206782, 'overall_f1': 0.599717613836922, 'overall_accuracy': 0.8161976047904191}
			------------EPOCH 29---------------
Loss:  tensor(0.0063, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0084, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0044, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0597, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1685, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0048, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0369, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0049, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0308, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0396, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0892, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0546, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0128, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0146, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0657, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0154, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1965, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5009615384615385, 'recall': 0.5189243027888446, 'f1': 0.5097847358121331, 'number': 1004}, 'P': {'precision': 0.5945121951219512, 'recall': 0.684610883557636, 'f1': 0.6363883600761491, 'number': 1709}, 'overall_precision': 0.5621675531914894, 'overall_recall': 0.6232952451161077, 'overall_f1': 0.5911553924139136, 'overall_accuracy': 0.8159730538922155}
			------------EPOCH 30---------------
Loss:  tensor(0.0058, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0074, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0036, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0566, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1629, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0041, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0308, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0043, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0231, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0365, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0785, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0555, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0131, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0156, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0602, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0141, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1827, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4893203883495146, 'recall': 0.50199203187251, 'f1': 0.4955752212389381, 'number': 1004}, 'P': {'precision': 0.5848012078510317, 'recall': 0.6799297834991223, 'f1': 0.6287878787878788, 'number': 1709}, 'overall_precision': 0.5522041763341067, 'overall_recall': 0.6140803538518246, 'overall_f1': 0.581500872600349, 'overall_accuracy': 0.8162125748502994}


		-------------RUN 3-----------
			------------EPOCH 1---------------
Loss:  tensor(3.2308, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.4362, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.4405, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.4766, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3577, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4069, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.9879, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.0865, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.0960, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4762, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6847, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.7637, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.9675, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.7797, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2716, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.7750, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5886, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1944, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6256, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5417, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3592, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1948, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.4814, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3113, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.3805, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.11171366594360087, 'recall': 0.08401305057096248, 'f1': 0.09590316573556797, 'number': 1226}, 'P': {'precision': 0.3404977375565611, 'recall': 0.4213719085394307, 'f1': 0.37664233576642336, 'number': 2143}, 'overall_precision': 0.28147733631785116, 'overall_recall': 0.2986049272781241, 'overall_f1': 0.2897882759614, 'overall_accuracy': 0.682101630370106}
			------------EPOCH 2---------------
Loss:  tensor(2.0786, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2055, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.1266, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.1543, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8970, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2859, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.9977, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.9994, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4346, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3382, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4627, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.7975, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2218, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.8805, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5192, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.7666, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.7487, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1442, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1069, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5347, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2435, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0976, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6725, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2197, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5627, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.2718351324828263, 'recall': 0.45187601957585644, 'f1': 0.3394607843137255, 'number': 1226}, 'P': {'precision': 0.30676328502415456, 'recall': 0.35557629491367243, 'f1': 0.3293710827750162, 'number': 2143}, 'overall_precision': 0.29102167182662536, 'overall_recall': 0.39062036212525975, 'overall_f1': 0.333544544417691, 'overall_accuracy': 0.7057241461382839}
			------------EPOCH 3---------------
Loss:  tensor(1.7037, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5853, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3693, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3639, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6296, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3360, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3442, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4607, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0694, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2476, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3705, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3723, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9818, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5664, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2177, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3270, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3813, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1125, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8502, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1620, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1903, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0682, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3731, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1517, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2355, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3452914798206278, 'recall': 0.4396411092985318, 'f1': 0.38679583781844273, 'number': 1226}, 'P': {'precision': 0.43836113125453224, 'recall': 0.564162389174055, 'f1': 0.49336870026525187, 'number': 2143}, 'overall_precision': 0.4047233155823107, 'overall_recall': 0.5188483229444939, 'overall_f1': 0.45473465140478664, 'overall_accuracy': 0.7730453409496608}
			------------EPOCH 4---------------
Loss:  tensor(1.3150, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2250, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0839, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0902, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4473, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2778, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1253, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7937, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1961, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2997, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0904, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7769, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2112, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9641, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9800, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1835, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0909, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6521, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0218, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1418, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0513, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9568, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1181, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9467, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3747412008281574, 'recall': 0.4429037520391517, 'f1': 0.40598130841121494, 'number': 1226}, 'P': {'precision': 0.4504442925495557, 'recall': 0.6150256649556696, 'f1': 0.5200236733083449, 'number': 2143}, 'overall_precision': 0.42537142857142857, 'overall_recall': 0.5523894330661917, 'overall_f1': 0.4806301652892562, 'overall_accuracy': 0.7867785314768535}
			------------EPOCH 5---------------
Loss:  tensor(1.1648, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0241, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8384, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3091, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2620, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8265, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8059, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5922, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1445, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3188, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8963, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5724, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8680, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7506, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7227, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9428, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0813, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4876, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7409, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1094, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0301, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6928, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0835, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7221, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.35139484978540775, 'recall': 0.534257748776509, 'f1': 0.42394822006472493, 'number': 1226}, 'P': {'precision': 0.47763457164518575, 'recall': 0.5879608026131591, 'f1': 0.527086383601757, 'number': 2143}, 'overall_precision': 0.4253665037760995, 'overall_recall': 0.568417928168596, 'overall_f1': 0.4865963664083344, 'overall_accuracy': 0.7929905985957396}
			------------EPOCH 6---------------
Loss:  tensor(1.0014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8864, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5847, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6181, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2415, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2278, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5619, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6032, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4244, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1156, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2498, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6217, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4298, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6891, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5816, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5665, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7932, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0466, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3417, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6123, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0996, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0164, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4664, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0518, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4957, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3361621876473362, 'recall': 0.5815660685154975, 'f1': 0.42605318195398867, 'number': 1226}, 'P': {'precision': 0.5486111111111112, 'recall': 0.589827344843677, 'f1': 0.568473127951428, 'number': 2143}, 'overall_precision': 0.44677966101694916, 'overall_recall': 0.5868210151380232, 'overall_f1': 0.5073133179368746, 'overall_accuracy': 0.7680828275615852}
			------------EPOCH 7---------------
Loss:  tensor(0.9545, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7914, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4411, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4452, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1700, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2206, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4138, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4775, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3364, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0827, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2151, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4684, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2906, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4799, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3995, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4341, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6308, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0222, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2623, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4742, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0750, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0196, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3338, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0311, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4293, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.45989304812834225, 'recall': 0.42088091353996737, 'f1': 0.43952299829642244, 'number': 1226}, 'P': {'precision': 0.5650529500756429, 'recall': 0.6971535230984601, 'f1': 0.6241905159807813, 'number': 2143}, 'overall_precision': 0.5337227827934148, 'overall_recall': 0.5966162065894924, 'overall_f1': 0.5634197617379118, 'overall_accuracy': 0.7916577412828751}
			------------EPOCH 8---------------
Loss:  tensor(0.9493, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8208, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4123, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2164, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1927, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7293, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5035, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2911, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0533, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2286, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4094, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4812, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7489, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4960, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5071, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6934, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0231, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1974, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4539, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0774, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0162, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6486, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0708, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6334, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4039039039039039, 'recall': 0.6582381729200653, 'f1': 0.500620347394541, 'number': 1226}, 'P': {'precision': 0.663337250293772, 'recall': 0.5268315445636957, 'f1': 0.587256176853056, 'number': 2143}, 'overall_precision': 0.5232432432432432, 'overall_recall': 0.574651231819531, 'overall_f1': 0.5477436695430754, 'overall_accuracy': 0.7802570510531953}
			------------EPOCH 9---------------
Loss:  tensor(0.9695, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6999, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3953, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3666, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1516, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1611, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4757, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5031, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4836, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0810, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2193, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5596, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2574, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4955, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2374, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4060, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4868, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0177, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1610, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3102, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0669, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0134, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6253, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0672, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5359, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3269054178145087, 'recall': 0.5807504078303426, 'f1': 0.41833137485311406, 'number': 1226}, 'P': {'precision': 0.46153846153846156, 'recall': 0.4255716285580961, 'f1': 0.44282592862345227, 'number': 2143}, 'overall_precision': 0.3909484833895041, 'overall_recall': 0.48204214900563963, 'overall_f1': 0.43174265585537686, 'overall_accuracy': 0.7548732595501606}
			------------EPOCH 10---------------
Loss:  tensor(0.8665, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8755, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3776, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3372, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1380, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0899, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1810, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2221, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0391, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1660, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3790, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2532, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5724, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3482, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5592, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8202, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0450, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2762, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4605, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0826, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0173, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2670, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0259, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1664, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.46369518332135157, 'recall': 0.5261011419249593, 'f1': 0.49293083683607186, 'number': 1226}, 'P': {'precision': 0.5682087781731909, 'recall': 0.6705552963135791, 'f1': 0.615154109589041, 'number': 2143}, 'overall_precision': 0.5311224489795918, 'overall_recall': 0.6179875333926982, 'overall_f1': 0.5712717793936068, 'overall_accuracy': 0.820564084255623}
			------------EPOCH 11---------------
Loss:  tensor(0.6493, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5161, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1514, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1528, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1116, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0865, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2715, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1617, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1937, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0434, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2459, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4384, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1952, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4076, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1778, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1925, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4132, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0090, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0564, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1353, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0515, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0041, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1665, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0187, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1766, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5585714285714286, 'recall': 0.3189233278955954, 'f1': 0.40602284527518173, 'number': 1226}, 'P': {'precision': 0.514367816091954, 'recall': 0.7517498833411106, 'f1': 0.6108056872037915, 'number': 2143}, 'overall_precision': 0.5224425887265136, 'overall_recall': 0.5942416147224696, 'overall_f1': 0.5560338841827525, 'overall_accuracy': 0.8045816970129716}
			------------EPOCH 12---------------
Loss:  tensor(0.7192, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5839, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2150, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1757, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0845, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0776, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0813, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0637, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1426, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0130, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1250, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1203, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3581, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1489, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2734, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5181, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0092, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0670, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2025, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0121, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0121, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0972, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0074, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0649, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4279411764705882, 'recall': 0.47471451876019577, 'f1': 0.4501160092807424, 'number': 1226}, 'P': {'precision': 0.5621909471281856, 'recall': 0.6896873541763883, 'f1': 0.6194467728415759, 'number': 2143}, 'overall_precision': 0.5164201554274254, 'overall_recall': 0.6114574057583853, 'overall_f1': 0.5599347648817614, 'overall_accuracy': 0.8109127692490777}
			------------EPOCH 13---------------
Loss:  tensor(0.4911, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3606, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0641, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0877, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0873, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0681, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1608, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1082, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1982, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0178, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1029, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1807, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0970, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3329, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0591, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0831, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3034, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0063, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0207, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0548, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0087, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0025, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0449, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0042, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0739, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.40788762830902214, 'recall': 0.6158238172920065, 'f1': 0.49073773155671113, 'number': 1226}, 'P': {'precision': 0.6107954545454546, 'recall': 0.6019598693420438, 'f1': 0.6063454759106933, 'number': 2143}, 'overall_precision': 0.5160232147363109, 'overall_recall': 0.6070050460077174, 'overall_f1': 0.5578286961265685, 'overall_accuracy': 0.8075925264786386}
			------------EPOCH 14---------------
Loss:  tensor(0.4745, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3573, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0884, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0693, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0734, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0543, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0602, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0349, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0665, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0056, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0909, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1219, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0432, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1652, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0398, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0414, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2725, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0136, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0371, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0075, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0355, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0026, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0404, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4749262536873156, 'recall': 0.5252854812398042, 'f1': 0.4988381099922541, 'number': 1226}, 'P': {'precision': 0.5930232558139535, 'recall': 0.6901539897340178, 'f1': 0.637912443390123, 'number': 2143}, 'overall_precision': 0.5514285714285714, 'overall_recall': 0.6301573167111902, 'overall_f1': 0.5881701066629728, 'overall_accuracy': 0.8218374390098774}
			------------EPOCH 15---------------
Loss:  tensor(0.4112, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2963, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0284, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0480, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0519, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0487, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0347, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0142, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0462, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0040, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0826, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0986, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0319, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1724, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0321, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0280, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2557, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0081, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0269, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0058, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0267, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0283, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4435028248587571, 'recall': 0.5122349102773246, 'f1': 0.4753974261922786, 'number': 1226}, 'P': {'precision': 0.594855305466238, 'recall': 0.6906206252916472, 'f1': 0.6391708054415893, 'number': 2143}, 'overall_precision': 0.5399590163934426, 'overall_recall': 0.6257049569605224, 'overall_f1': 0.5796782620651726, 'overall_accuracy': 0.817446150184458}
			------------EPOCH 16---------------
Loss:  tensor(0.3841, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2809, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0201, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0205, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0479, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0445, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0217, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0104, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0560, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0031, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0772, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0849, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0251, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1848, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0253, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0208, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2387, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0054, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0193, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0031, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0237, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0205, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.46842105263157896, 'recall': 0.5081566068515497, 'f1': 0.4874804381846635, 'number': 1226}, 'P': {'precision': 0.5925779707856297, 'recall': 0.7004199720018666, 'f1': 0.6420017108639863, 'number': 2143}, 'overall_precision': 0.5498317369919752, 'overall_recall': 0.6304541406945681, 'overall_f1': 0.5873893805309734, 'overall_accuracy': 0.820611686302511}
			------------EPOCH 17---------------
Loss:  tensor(0.3654, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2460, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0162, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0102, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0490, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0416, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0152, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0083, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0732, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0710, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0760, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0207, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1555, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0203, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0156, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2283, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0043, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0159, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0211, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0128, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.438179347826087, 'recall': 0.5261011419249593, 'f1': 0.4781319495922906, 'number': 1226}, 'P': {'precision': 0.5967346938775511, 'recall': 0.6822211852543164, 'f1': 0.6366209449161767, 'number': 2143}, 'overall_precision': 0.5372259051504334, 'overall_recall': 0.6254081329771446, 'overall_f1': 0.5779728432313811, 'overall_accuracy': 0.8195763417826967}
			------------EPOCH 18---------------
Loss:  tensor(0.3514, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2318, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0137, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0074, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0431, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0408, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0156, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0068, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0330, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0698, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0725, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0171, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1199, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0173, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0128, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2110, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0037, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0135, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0188, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0111, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.43085787451984636, 'recall': 0.5489396411092985, 'f1': 0.4827833572453371, 'number': 1226}, 'P': {'precision': 0.6073853989813243, 'recall': 0.6677554829678022, 'f1': 0.6361413647477216, 'number': 2143}, 'overall_precision': 0.5370086778968862, 'overall_recall': 0.624517661027011, 'overall_f1': 0.5774667215589406, 'overall_accuracy': 0.8181958824229442}
			------------EPOCH 19---------------
Loss:  tensor(0.3454, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2195, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0123, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0060, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0414, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0392, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0144, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0056, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0396, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0660, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0669, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0141, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1337, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0138, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0106, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1935, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0031, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0119, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0158, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0093, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.47098214285714285, 'recall': 0.5163132137030995, 'f1': 0.4926070038910506, 'number': 1226}, 'P': {'precision': 0.5925632911392406, 'recall': 0.6990200653289781, 'f1': 0.6414044101905374, 'number': 2143}, 'overall_precision': 0.5503615702479339, 'overall_recall': 0.6325319085782131, 'overall_f1': 0.588592735809971, 'overall_accuracy': 0.8229203855765798}
			------------EPOCH 20---------------
Loss:  tensor(0.3361, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2126, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0100, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0364, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0378, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0129, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0045, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0377, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0566, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0568, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0116, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1172, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0085, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0102, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1951, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0110, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0116, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0088, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.43209876543209874, 'recall': 0.5424143556280587, 'f1': 0.4810126582278481, 'number': 1226}, 'P': {'precision': 0.5931291390728477, 'recall': 0.6686887540830612, 'f1': 0.6286466330335599, 'number': 2143}, 'overall_precision': 0.5304677623261694, 'overall_recall': 0.6227367171267438, 'overall_f1': 0.5729109776078646, 'overall_accuracy': 0.817743662977508}
			------------EPOCH 21---------------
Loss:  tensor(0.3170, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2046, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0094, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0044, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0338, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0359, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0116, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0040, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0372, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0783, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0628, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0101, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0956, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0055, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0095, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2115, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0025, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0092, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0110, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0082, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.45113788487282463, 'recall': 0.5497553017944535, 'f1': 0.49558823529411766, 'number': 1226}, 'P': {'precision': 0.6147926464300983, 'recall': 0.6710219318712086, 'f1': 0.6416778224007139, 'number': 2143}, 'overall_precision': 0.5510044351682755, 'overall_recall': 0.6268922528940338, 'overall_f1': 0.5865037489586226, 'overall_accuracy': 0.8184814947042723}
			------------EPOCH 22---------------
Loss:  tensor(0.3227, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1877, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0093, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0045, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0312, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0355, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0103, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0039, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0302, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0704, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0667, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0082, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0932, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0048, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0119, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1978, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0094, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0106, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0075, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.46881878209831257, 'recall': 0.5212071778140294, 'f1': 0.4936268829663963, 'number': 1226}, 'P': {'precision': 0.5941623350659736, 'recall': 0.6934204386374242, 'f1': 0.6399655469422911, 'number': 2143}, 'overall_precision': 0.5499482401656315, 'overall_recall': 0.630750964677946, 'overall_f1': 0.58758468132172, 'overall_accuracy': 0.8228965845531357}
			------------EPOCH 23---------------
Loss:  tensor(0.3091, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1743, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0087, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0300, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0322, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0153, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0036, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0292, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0711, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0804, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0071, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0942, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0039, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0192, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1706, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0072, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0098, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0069, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.44368600682593856, 'recall': 0.5301794453507341, 'f1': 0.4830917874396135, 'number': 1226}, 'P': {'precision': 0.5917039446929646, 'recall': 0.67895473635091, 'f1': 0.6323337679269883, 'number': 2143}, 'overall_precision': 0.5364424057084608, 'overall_recall': 0.6248144850103888, 'overall_f1': 0.5772658713835184, 'overall_accuracy': 0.8197548494585267}
			------------EPOCH 24---------------
Loss:  tensor(0.2970, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1831, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0075, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0033, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0275, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0331, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0089, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0030, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0265, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0514, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0505, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0065, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0869, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0034, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0091, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1654, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0067, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0102, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0071, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.42479018721755973, 'recall': 0.5367047308319739, 'f1': 0.4742342342342343, 'number': 1226}, 'P': {'precision': 0.5899221630479312, 'recall': 0.6719552029864676, 'f1': 0.6282722513089005, 'number': 2143}, 'overall_precision': 0.5258145363408522, 'overall_recall': 0.6227367171267438, 'overall_f1': 0.5701861665987226, 'overall_accuracy': 0.8181839819112222}
			------------EPOCH 25---------------
Loss:  tensor(0.3224, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2377, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0116, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0031, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0277, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0302, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0081, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0028, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0263, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0466, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0601, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0063, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0952, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0031, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0069, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1440, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0066, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0090, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0062, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.47176554681915656, 'recall': 0.5383360522022839, 'f1': 0.5028571428571429, 'number': 1226}, 'P': {'precision': 0.5981270358306189, 'recall': 0.6854876341577228, 'f1': 0.6388345292454881, 'number': 2143}, 'overall_precision': 0.5522697795071336, 'overall_recall': 0.6319382606114574, 'overall_f1': 0.5894241417497231, 'overall_accuracy': 0.820504581697013}
			------------EPOCH 26---------------
Loss:  tensor(0.2757, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1554, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0067, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0029, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0242, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0299, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0072, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0026, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0272, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0410, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0365, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0053, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0961, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0026, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0056, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1157, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0056, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0084, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0057, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4245343609505459, 'recall': 0.5391517128874388, 'f1': 0.47502694933524975, 'number': 1226}, 'P': {'precision': 0.5885047896709704, 'recall': 0.6593560429304713, 'f1': 0.621919014084507, 'number': 2143}, 'overall_precision': 0.5240020212228398, 'overall_recall': 0.6156129415256753, 'overall_f1': 0.5661252900232018, 'overall_accuracy': 0.8161132928715935}
			------------EPOCH 27---------------
Loss:  tensor(0.2468, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1181, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0059, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0025, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0224, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0283, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0300, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0340, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0321, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0043, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1032, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0025, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1141, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0052, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0080, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0054, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.46685472496473907, 'recall': 0.5399673735725938, 'f1': 0.5007564296520424, 'number': 1226}, 'P': {'precision': 0.5986059860598606, 'recall': 0.6812879141390574, 'f1': 0.637276298559581, 'number': 2143}, 'overall_precision': 0.5501685247601763, 'overall_recall': 0.6298604927278124, 'overall_f1': 0.5873235538333795, 'overall_accuracy': 0.8216351303106033}
			------------EPOCH 28---------------
Loss:  tensor(0.2311, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1089, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0052, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0241, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0264, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0051, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0315, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0480, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0420, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0030, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0966, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1283, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0050, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0073, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0054, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.453416149068323, 'recall': 0.535889070146819, 'f1': 0.4912149532710281, 'number': 1226}, 'P': {'precision': 0.579094913896676, 'recall': 0.6747550163322446, 'f1': 0.6232758620689656, 'number': 2143}, 'overall_precision': 0.5329447541814496, 'overall_recall': 0.6242208370436332, 'overall_f1': 0.5749829118250172, 'overall_accuracy': 0.8190289182434844}
			------------EPOCH 29---------------
Loss:  tensor(0.2132, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1164, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0071, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0175, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0312, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0048, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0307, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0310, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0296, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0875, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0034, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0913, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0043, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0071, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0050, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4621790423317141, 'recall': 0.5432300163132137, 'f1': 0.49943757030371194, 'number': 1226}, 'P': {'precision': 0.6056338028169014, 'recall': 0.6822211852543164, 'f1': 0.6416502084704849, 'number': 2143}, 'overall_precision': 0.5520103761348898, 'overall_recall': 0.6316414366280796, 'overall_f1': 0.5891472868217055, 'overall_accuracy': 0.8209806021658931}
			------------EPOCH 30---------------
Loss:  tensor(0.2030, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0885, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0045, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0156, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0256, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0047, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0260, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0276, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0235, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0854, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0033, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0953, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0041, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0059, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0044, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4572619874913134, 'recall': 0.5367047308319739, 'f1': 0.4938086303939962, 'number': 1226}, 'P': {'precision': 0.5842105263157895, 'recall': 0.6733551096593561, 'f1': 0.6256232386733146, 'number': 2143}, 'overall_precision': 0.537477615758506, 'overall_recall': 0.6236271890768774, 'overall_f1': 0.5773564165979664, 'overall_accuracy': 0.8195763417826967}


		-------------RUN 4-----------
			------------EPOCH 1---------------
Loss:  tensor(2.3607, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0981, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.9002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.0176, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.8736, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.8840, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2171, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.6611, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.5680, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0220, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2096, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.3533, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2931, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.4525, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2897, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2143, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.9934, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.1404, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.9710, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.1687, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1572, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.0755, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3638, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4307, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0350, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.15853658536585366, 'recall': 0.01145374449339207, 'f1': 0.021364009860312245, 'number': 1135}, 'P': {'precision': 0.3598014888337469, 'recall': 0.4452405322415558, 'f1': 0.3979871912168344, 'number': 1954}, 'overall_precision': 0.3532, 'overall_recall': 0.28585302686953706, 'overall_f1': 0.3159778135623546, 'overall_accuracy': 0.7100916230366492}
			------------EPOCH 2---------------
Loss:  tensor(1.3170, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6340, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1237, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.7250, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.7680, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.8267, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1048, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6210, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6378, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6700, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.7401, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6303, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6331, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6674, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1920, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4831, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4363, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5719, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6848, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1178, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3864, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9698, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9671, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8175, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3155588020452885, 'recall': 0.38061674008810575, 'f1': 0.34504792332268375, 'number': 1135}, 'P': {'precision': 0.5491606714628298, 'recall': 0.5859774820880246, 'f1': 0.5669720227779154, 'number': 1954}, 'overall_precision': 0.45657209033005214, 'overall_recall': 0.5105212042732276, 'overall_f1': 0.48204187681491667, 'overall_accuracy': 0.7836387434554973}
			------------EPOCH 3---------------
Loss:  tensor(0.9280, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4525, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8113, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2491, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3139, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3612, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0710, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2126, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1398, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4910, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3703, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1620, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1741, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2323, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1443, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1630, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2783, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2516, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0775, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1760, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0752, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9808, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7870, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7211, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6196, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3933933933933934, 'recall': 0.4616740088105727, 'f1': 0.4248074584515606, 'number': 1135}, 'P': {'precision': 0.5393360618462938, 'recall': 0.6069600818833163, 'f1': 0.5711533830965568, 'number': 1954}, 'overall_precision': 0.4842820730671198, 'overall_recall': 0.5535772094528973, 'overall_f1': 0.5166163141993957, 'overall_accuracy': 0.8033900523560209}
			------------EPOCH 4---------------
Loss:  tensor(0.6759, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3390, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5850, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9380, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0463, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0448, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9045, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8110, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3781, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1641, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9663, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9132, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8957, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0853, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8530, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0191, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0307, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8510, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8631, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0528, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7371, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7324, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5457, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4535, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4082018927444795, 'recall': 0.5700440528634361, 'f1': 0.47573529411764703, 'number': 1135}, 'P': {'precision': 0.6219825372367745, 'recall': 0.6197543500511771, 'f1': 0.6208664445014099, 'number': 1954}, 'overall_precision': 0.5260475651189128, 'overall_recall': 0.6014891550663645, 'overall_f1': 0.5612445249962241, 'overall_accuracy': 0.8152748691099476}
			------------EPOCH 5---------------
Loss:  tensor(0.4799, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2395, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4352, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6991, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6767, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7181, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0278, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6683, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5808, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2970, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9805, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8125, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7484, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7553, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0662, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6039, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8581, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8593, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7093, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6417, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0328, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5303, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5304, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3609, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3470, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4593781344032096, 'recall': 0.4035242290748899, 'f1': 0.42964352720450283, 'number': 1135}, 'P': {'precision': 0.6081196581196581, 'recall': 0.7282497441146366, 'f1': 0.6627852817885421, 'number': 1954}, 'overall_precision': 0.563679952052742, 'overall_recall': 0.6089349303981871, 'overall_f1': 0.5854341736694678, 'overall_accuracy': 0.8236387434554974}
			------------EPOCH 6---------------
Loss:  tensor(0.4643, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2172, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3278, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5160, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4425, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5531, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0313, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5431, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6587, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2862, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8571, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6920, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6111, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5044, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0505, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4285, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0408, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7968, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5144, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4867, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0297, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4710, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6253, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4095, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2636, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3771289537712895, 'recall': 0.5462555066079295, 'f1': 0.4462036703850306, 'number': 1135}, 'P': {'precision': 0.6352097130242825, 'recall': 0.5890481064483112, 'f1': 0.6112586298459904, 'number': 1954}, 'overall_precision': 0.5124421296296297, 'overall_recall': 0.5733247005503399, 'overall_f1': 0.5411764705882353, 'overall_accuracy': 0.8016753926701571}
			------------EPOCH 7---------------
Loss:  tensor(0.3203, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1500, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1810, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3210, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3777, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3767, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0134, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4162, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3409, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1661, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8095, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6816, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5934, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4107, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0338, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4077, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7083, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6426, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5167, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5474, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0225, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3791, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4637, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2157, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1775, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.38626907073509015, 'recall': 0.4907488986784141, 'f1': 0.4322856034148234, 'number': 1135}, 'P': {'precision': 0.5373786407766991, 'recall': 0.5665301944728761, 'f1': 0.5515695067264574, 'number': 1954}, 'overall_precision': 0.47515705311250717, 'overall_recall': 0.5386856587892522, 'overall_f1': 0.504930966469428, 'overall_accuracy': 0.7954057591623037}
			------------EPOCH 8---------------
Loss:  tensor(0.2219, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1096, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2192, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3850, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4110, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4444, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0128, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3894, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2594, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1337, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6725, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5063, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4440, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2614, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0257, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3390, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6708, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6642, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4693, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4996, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0259, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3640, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4482, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2374, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2008, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.47474747474747475, 'recall': 0.4555066079295154, 'f1': 0.46492805755395683, 'number': 1135}, 'P': {'precision': 0.5358666149670415, 'recall': 0.7072671443193449, 'f1': 0.6097507169644826, 'number': 1954}, 'overall_precision': 0.5177208287895311, 'overall_recall': 0.6147620589187439, 'overall_f1': 0.5620837649844607, 'overall_accuracy': 0.827761780104712}
			------------EPOCH 9---------------
Loss:  tensor(0.2018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1547, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1039, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1953, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2270, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1939, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0216, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4412, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3298, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1474, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8217, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7424, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4550, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3629, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0234, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3193, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5779, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5278, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3522, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0160, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2777, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4882, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2357, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3461, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.522867737948084, 'recall': 0.3726872246696035, 'f1': 0.43518518518518523, 'number': 1135}, 'P': {'precision': 0.5368837711406981, 'recall': 0.7635619242579325, 'f1': 0.6304669342911472, 'number': 1954}, 'overall_precision': 0.5337235228539576, 'overall_recall': 0.6199417287147945, 'overall_f1': 0.5736109031001947, 'overall_accuracy': 0.8136910994764398}
			------------EPOCH 10---------------
Loss:  tensor(0.3653, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1689, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2319, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2615, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2163, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1729, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0118, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2519, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2420, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1157, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8455, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6806, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6128, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4643, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0403, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4145, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5921, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5607, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3391, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2732, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0103, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1671, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4563, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1357, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2112, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.6086956521739131, 'recall': 0.24669603524229075, 'f1': 0.3510971786833856, 'number': 1135}, 'P': {'precision': 0.5750096413420748, 'recall': 0.763050153531218, 'f1': 0.6558170222124476, 'number': 1954}, 'overall_precision': 0.5800851621356043, 'overall_recall': 0.5733247005503399, 'overall_f1': 0.5766851188537935, 'overall_accuracy': 0.7889528795811518}
			------------EPOCH 11---------------
Loss:  tensor(0.2446, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1444, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2387, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6575, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4074, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0069, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4814, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1574, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0690, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5966, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4339, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2977, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2046, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0438, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4422, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9242, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9163, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7851, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6998, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0180, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4102, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3438, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1466, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1418, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.40786516853932586, 'recall': 0.3198237885462555, 'f1': 0.3585185185185185, 'number': 1135}, 'P': {'precision': 0.45, 'recall': 0.6586489252814739, 'f1': 0.53469048608226, 'number': 1954}, 'overall_precision': 0.44, 'overall_recall': 0.534153447717708, 'overall_f1': 0.48252668518789293, 'overall_accuracy': 0.8095811518324607}
			------------EPOCH 12---------------
Loss:  tensor(0.1594, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0837, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1110, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1893, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3804, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2280, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0116, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6738, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3251, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1865, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0158, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7462, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4424, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2355, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0143, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2428, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5294, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4586, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3219, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3548, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0139, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3111, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4348, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2321, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1818, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4118279569892473, 'recall': 0.6748898678414097, 'f1': 0.5115191986644406, 'number': 1135}, 'P': {'precision': 0.6955128205128205, 'recall': 0.5552712384851587, 'f1': 0.6175298804780877, 'number': 1954}, 'overall_precision': 0.5412280701754386, 'overall_recall': 0.5992230495305925, 'overall_f1': 0.5687509602089416, 'overall_accuracy': 0.7955104712041885}
			------------EPOCH 13---------------
Loss:  tensor(0.2050, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1239, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0965, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3581, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1808, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1811, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0166, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2267, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1448, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0518, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5276, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2377, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1049, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0154, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1901, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4893, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3780, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2560, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2134, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0102, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0926, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2525, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0901, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0808, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.548618219037871, 'recall': 0.4722466960352423, 'f1': 0.5075757575757575, 'number': 1135}, 'P': {'precision': 0.6165695253955037, 'recall': 0.7579324462640737, 'f1': 0.679981634527089, 'number': 1954}, 'overall_precision': 0.5969221663213968, 'overall_recall': 0.6529621236646164, 'overall_f1': 0.6236858379715522, 'overall_accuracy': 0.8357329842931938}
			------------EPOCH 14---------------
Loss:  tensor(0.1015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0429, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0340, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0775, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0718, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0454, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0048, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1513, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0994, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0252, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4626, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3598, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1809, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0762, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0084, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1409, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4129, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3102, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1622, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1498, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0039, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0446, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2342, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0427, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0602, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.498812351543943, 'recall': 0.5550660792951542, 'f1': 0.5254378648874062, 'number': 1135}, 'P': {'precision': 0.6561320754716982, 'recall': 0.7118730808597749, 'f1': 0.6828669612174767, 'number': 1954}, 'overall_precision': 0.5973987584983742, 'overall_recall': 0.654257041113629, 'overall_f1': 0.6245364647713226, 'overall_accuracy': 0.8340183246073298}
			------------EPOCH 15---------------
Loss:  tensor(0.0528, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0332, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0173, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0574, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0443, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0234, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0033, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0553, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0119, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4281, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3196, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1520, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0356, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1182, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3824, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2406, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1518, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1273, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0310, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2126, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0297, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0496, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.519327731092437, 'recall': 0.5444933920704845, 'f1': 0.5316129032258065, 'number': 1135}, 'P': {'precision': 0.6480036714089031, 'recall': 0.7226202661207779, 'f1': 0.6832809097507863, 'number': 1954}, 'overall_precision': 0.6025526862570496, 'overall_recall': 0.6571706053739074, 'overall_f1': 0.6286776091669247, 'overall_accuracy': 0.8345026178010472}
			------------EPOCH 16---------------
Loss:  tensor(0.0461, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0227, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0118, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0317, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0327, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0166, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0029, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0907, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0375, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0080, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4251, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3235, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1405, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0280, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1105, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3680, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2052, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1053, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1281, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0207, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2045, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0224, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0405, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.47878787878787876, 'recall': 0.5568281938325991, 'f1': 0.5148676171079429, 'number': 1135}, 'P': {'precision': 0.6490384615384616, 'recall': 0.6908904810644831, 'f1': 0.6693108577094696, 'number': 1954}, 'overall_precision': 0.5829411764705882, 'overall_recall': 0.6416315959857559, 'overall_f1': 0.6108799506857759, 'overall_accuracy': 0.8307722513089005}
			------------EPOCH 17---------------
Loss:  tensor(0.0426, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0053, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0091, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0208, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0248, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0129, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0708, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0287, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0056, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4200, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3112, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1335, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0197, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0977, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3551, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1988, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0916, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1361, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0159, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1975, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0192, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0380, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5081685296646603, 'recall': 0.520704845814978, 'f1': 0.5143603133159268, 'number': 1135}, 'P': {'precision': 0.6275395033860045, 'recall': 0.7113613101330604, 'f1': 0.6668265771168146, 'number': 1954}, 'overall_precision': 0.5864416814683244, 'overall_recall': 0.6413078666235027, 'overall_f1': 0.6126488325344053, 'overall_accuracy': 0.8302748691099476}
			------------EPOCH 18---------------
Loss:  tensor(0.0408, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0033, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0063, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0180, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0228, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0097, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0869, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0138, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0063, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4076, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2976, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1305, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0183, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0903, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3423, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1610, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0743, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1043, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0124, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1824, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0134, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0299, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4837962962962963, 'recall': 0.5524229074889868, 'f1': 0.5158371040723981, 'number': 1135}, 'P': {'precision': 0.6582646631119728, 'recall': 0.6949846468781986, 'f1': 0.6761264625342295, 'number': 1954}, 'overall_precision': 0.5909496874069664, 'overall_recall': 0.6426027840725154, 'overall_f1': 0.6156947890818859, 'overall_accuracy': 0.829738219895288}
			------------EPOCH 19---------------
Loss:  tensor(0.0388, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0029, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0055, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0148, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0186, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0072, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0576, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0106, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0042, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3951, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2957, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1231, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0150, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0796, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3440, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1539, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0623, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1146, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0102, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1731, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0148, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0330, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.5175131348511384, 'recall': 0.520704845814978, 'f1': 0.5191040843214757, 'number': 1135}, 'P': {'precision': 0.620597414177441, 'recall': 0.7123848515864892, 'f1': 0.6633309506790565, 'number': 1954}, 'overall_precision': 0.5858197932053176, 'overall_recall': 0.641955325348009, 'overall_f1': 0.6126042632066729, 'overall_accuracy': 0.8309162303664921}
			------------EPOCH 20---------------
Loss:  tensor(0.0332, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0044, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0136, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0176, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0059, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0625, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0097, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0039, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3912, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1176, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0133, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0810, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3226, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1394, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0586, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0969, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0085, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1616, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0096, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0263, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.47435897435897434, 'recall': 0.5541850220264317, 'f1': 0.5111743193823648, 'number': 1135}, 'P': {'precision': 0.6478599221789884, 'recall': 0.6816786079836233, 'f1': 0.6643391521197008, 'number': 1954}, 'overall_precision': 0.5798344175044352, 'overall_recall': 0.6348332793784396, 'overall_f1': 0.6060887034461443, 'overall_accuracy': 0.827761780104712}
			------------EPOCH 21---------------
Loss:  tensor(0.0344, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0032, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0125, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0142, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0048, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0461, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0084, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0026, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3999, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2811, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1043, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0106, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0764, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3301, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1801, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0732, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0960, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0079, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1680, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0099, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0257, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.49413735343383586, 'recall': 0.5198237885462555, 'f1': 0.5066552168312581, 'number': 1135}, 'P': {'precision': 0.6213202497769849, 'recall': 0.7128966223132037, 'f1': 0.6639656816015252, 'number': 1954}, 'overall_precision': 0.5771245634458673, 'overall_recall': 0.641955325348009, 'overall_f1': 0.607816091954023, 'overall_accuracy': 0.8341492146596858}
			------------EPOCH 22---------------
Loss:  tensor(0.0360, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0030, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0121, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0135, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0045, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0592, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0087, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3602, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2494, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1033, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0117, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0717, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1313, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0622, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0899, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0075, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1567, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0062, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0228, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4980237154150198, 'recall': 0.5550660792951542, 'f1': 0.5249999999999999, 'number': 1135}, 'P': {'precision': 0.6517943743937924, 'recall': 0.6878198567041965, 'f1': 0.6693227091633467, 'number': 1954}, 'overall_precision': 0.5933273219116321, 'overall_recall': 0.6390417610877307, 'overall_f1': 0.6153366583541147, 'overall_accuracy': 0.8309162303664921}
			------------EPOCH 23---------------
Loss:  tensor(0.0332, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0107, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0101, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0041, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0427, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0078, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3381, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2422, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0874, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0106, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0616, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2747, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1190, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0469, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0917, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0126, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1499, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0054, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0193, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.45989304812834225, 'recall': 0.5303964757709251, 'f1': 0.49263502454991814, 'number': 1135}, 'P': {'precision': 0.6175663311985361, 'recall': 0.6908904810644831, 'f1': 0.6521739130434783, 'number': 1954}, 'overall_precision': 0.5585121602288984, 'overall_recall': 0.6319197151181613, 'overall_f1': 0.5929526123936817, 'overall_accuracy': 0.8317539267015707}
			------------EPOCH 24---------------
Loss:  tensor(0.0277, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0108, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0088, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0036, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0541, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0074, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4059, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2567, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0933, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0093, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0623, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2920, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1114, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0525, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0075, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1678, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0057, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0495, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.49362041467304624, 'recall': 0.5453744493392071, 'f1': 0.5182084554206782, 'number': 1135}, 'P': {'precision': 0.6492071119654013, 'recall': 0.6914022517911975, 'f1': 0.6696406443618339, 'number': 1954}, 'overall_precision': 0.5907046476761619, 'overall_recall': 0.637746843638718, 'overall_f1': 0.6133250311332502, 'overall_accuracy': 0.8292931937172775}
			------------EPOCH 25---------------
Loss:  tensor(0.0332, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0026, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0030, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0134, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0090, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0045, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0425, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0085, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3612, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2239, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0918, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0092, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0680, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2806, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1172, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0472, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1225, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0055, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1394, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0042, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0145, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.48607796340493237, 'recall': 0.5383259911894274, 'f1': 0.5108695652173912, 'number': 1135}, 'P': {'precision': 0.6262766945218199, 'recall': 0.6903787103377687, 'f1': 0.6567672833495619, 'number': 1954}, 'overall_precision': 0.574611550864849, 'overall_recall': 0.6345095500161865, 'overall_f1': 0.603076923076923, 'overall_accuracy': 0.829175392670157}
			------------EPOCH 26---------------
Loss:  tensor(0.0283, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0094, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0064, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0034, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0259, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0068, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2891, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1926, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0685, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0091, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0481, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2405, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0862, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0352, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0820, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0090, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1188, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0031, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0101, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.49599358974358976, 'recall': 0.5453744493392071, 'f1': 0.5195132186319765, 'number': 1135}, 'P': {'precision': 0.6431593794076164, 'recall': 0.7001023541453428, 'f1': 0.6704239157069345, 'number': 1954}, 'overall_precision': 0.5887407407407408, 'overall_recall': 0.6432502427970217, 'overall_f1': 0.6147896039603961, 'overall_accuracy': 0.8336518324607329}
			------------EPOCH 27---------------
Loss:  tensor(0.0245, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0089, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0057, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0026, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0248, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0063, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2604, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1902, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0590, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0080, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0459, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2135, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0714, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0271, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0776, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0033, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1094, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0026, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0095, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.48699763593380613, 'recall': 0.5444933920704845, 'f1': 0.5141430948419301, 'number': 1135}, 'P': {'precision': 0.6303738317757009, 'recall': 0.6903787103377687, 'f1': 0.6590131900341963, 'number': 1954}, 'overall_precision': 0.5770020533880903, 'overall_recall': 0.6367756555519586, 'overall_f1': 0.605417051400431, 'overall_accuracy': 0.8329712041884817}
			------------EPOCH 28---------------
Loss:  tensor(0.0231, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0081, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0043, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0188, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0060, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2645, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1755, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0532, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0078, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0382, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2025, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0654, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0235, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0749, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0029, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0992, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0120, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.49059561128526646, 'recall': 0.5515418502202644, 'f1': 0.5192866030692659, 'number': 1135}, 'P': {'precision': 0.6472569778633301, 'recall': 0.6883316274309109, 'f1': 0.6671626984126985, 'number': 1954}, 'overall_precision': 0.5876565295169947, 'overall_recall': 0.6380705730009711, 'overall_f1': 0.6118267887629987, 'overall_accuracy': 0.8303141361256544}
			------------EPOCH 29---------------
Loss:  tensor(0.0209, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0079, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0036, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0167, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0057, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2403, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1645, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0457, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0072, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0353, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1972, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0724, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0242, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0750, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0025, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0899, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0079, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4932162809257781, 'recall': 0.5444933920704845, 'f1': 0.5175879396984924, 'number': 1135}, 'P': {'precision': 0.6279500231374364, 'recall': 0.6944728761514841, 'f1': 0.6595382746051032, 'number': 1954}, 'overall_precision': 0.5785002929115407, 'overall_recall': 0.6393654904499838, 'overall_f1': 0.6074119637090574, 'overall_accuracy': 0.8321596858638743}
			------------EPOCH 30---------------
Loss:  tensor(0.0202, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0076, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0028, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0150, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0055, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2352, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1525, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0492, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0072, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0304, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1837, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0549, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0201, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0770, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0840, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0069, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4740740740740741, 'recall': 0.5638766519823789, 'f1': 0.5150905432595573, 'number': 1135}, 'P': {'precision': 0.6433805569125549, 'recall': 0.6740020470829069, 'f1': 0.6583354161459635, 'number': 1954}, 'overall_precision': 0.5760965557845158, 'overall_recall': 0.633538361929427, 'overall_f1': 0.6034535923527597, 'overall_accuracy': 0.8273298429319371}


		-------------RUN 5-----------
			------------EPOCH 1---------------
Loss:  tensor(2.6495, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.6721, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.2119, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2973, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4359, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7199, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(3.0506, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.9385, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.9951, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.1886, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.8972, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3870, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.7682, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.8161, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.6193, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.6774, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6603, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.1824, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.4913, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8400, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.3149, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.4562, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4016, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.1956521739130435, 'recall': 0.007358953393295176, 'f1': 0.014184397163120569, 'number': 1223}, 'P': {'precision': 0.3130681818181818, 'recall': 0.5159176029962547, 'f1': 0.3896746817538897, 'number': 2136}, 'overall_precision': 0.3115535614133483, 'overall_recall': 0.3307532003572492, 'overall_f1': 0.32086642599277976, 'overall_accuracy': 0.6151336473323922}
			------------EPOCH 2---------------
Loss:  tensor(1.6892, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.3158, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2357, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1703, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3064, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5029, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.2201, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.0928, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.1583, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5311, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.9845, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3155, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.9336, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1455, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8890, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(2.1019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.7602, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.8535, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5077, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6025, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6964, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6053, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6744, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.9542, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2899, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.2665964172813488, 'recall': 0.20686835650040883, 'f1': 0.23296500920810312, 'number': 1223}, 'P': {'precision': 0.5032430370087753, 'recall': 0.6175093632958801, 'f1': 0.5545511877233551, 'number': 2136}, 'overall_precision': 0.4403361344537815, 'overall_recall': 0.467996427508187, 'overall_f1': 0.453745129167268, 'overall_accuracy': 0.7293003832054005}
			------------EPOCH 3---------------
Loss:  tensor(1.2888, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6336, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6216, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1037, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2260, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3863, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.5926, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4606, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6795, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0764, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4466, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3078, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3974, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1265, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6403, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.6178, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2456, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3455, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4568, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2483, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2399, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4342, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2412, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.4422, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2336, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3037974683544304, 'recall': 0.2747342600163532, 'f1': 0.28853585229712325, 'number': 1223}, 'P': {'precision': 0.4916221033868093, 'recall': 0.6455992509363296, 'f1': 0.558186601902449, 'number': 2136}, 'overall_precision': 0.43850677576067504, 'overall_recall': 0.5105686216135755, 'overall_f1': 0.47180192572214585, 'overall_accuracy': 0.7569789652267793}
			------------EPOCH 4---------------
Loss:  tensor(1.0626, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3075, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1340, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0782, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1682, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2932, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.2741, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1570, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.3687, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9101, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1294, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3078, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0358, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0890, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4619, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1873, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0199, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3304, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9722, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9068, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3287, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9710, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1239, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1607, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.36195286195286197, 'recall': 0.3515944399018806, 'f1': 0.3566984653670676, 'number': 1223}, 'P': {'precision': 0.5132969034608379, 'recall': 0.6596441947565543, 'f1': 0.577340708871133, 'number': 2136}, 'overall_precision': 0.467581998474447, 'overall_recall': 0.547484370348318, 'overall_f1': 0.5043883708173341, 'overall_accuracy': 0.7856303906796853}
			------------EPOCH 5---------------
Loss:  tensor(0.7446, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9850, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8499, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0441, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1414, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2328, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9949, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.8753, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0516, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6973, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2803, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6900, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0610, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3336, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7575, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5725, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6834, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2966, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6834, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6432, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2118, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7476, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9526, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1253, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.37444279346210996, 'recall': 0.41210139002452983, 'f1': 0.39237057220708443, 'number': 1223}, 'P': {'precision': 0.5427580893682589, 'recall': 0.6596441947565543, 'f1': 0.595519864750634, 'number': 2136}, 'overall_precision': 0.48528665651953323, 'overall_recall': 0.5695147365287287, 'overall_f1': 0.5240378030406793, 'overall_accuracy': 0.7990485116681892}
			------------EPOCH 6---------------
Loss:  tensor(0.4914, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6956, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5953, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0175, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1115, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1870, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7275, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6830, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7995, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4681, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6082, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2441, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4885, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0451, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2459, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4606, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3547, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4128, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2375, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4463, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4881, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1677, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4438, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7061, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1310, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3162356930902925, 'recall': 0.6099754701553557, 'f1': 0.4165270798436627, 'number': 1223}, 'P': {'precision': 0.5480489225393127, 'recall': 0.4405430711610487, 'f1': 0.48845055800674797, 'number': 2136}, 'overall_precision': 0.4138861629048086, 'overall_recall': 0.5022328073831498, 'overall_f1': 0.4537995965030262, 'overall_accuracy': 0.7474522179643844}
			------------EPOCH 7---------------
Loss:  tensor(0.6132, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.0232, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4616, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0161, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0804, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1448, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7199, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6991, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(1.1091, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4668, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4604, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2045, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3738, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0355, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3449, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7029, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6992, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.9368, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3549, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5676, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3269, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1163, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3456, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6063, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1163, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.441747572815534, 'recall': 0.223221586263287, 'f1': 0.2965779467680608, 'number': 1223}, 'P': {'precision': 0.48273736128236744, 'recall': 0.7331460674157303, 'f1': 0.5821561338289963, 'number': 2136}, 'overall_precision': 0.4761781460383221, 'overall_recall': 0.547484370348318, 'overall_f1': 0.5093477357706689, 'overall_accuracy': 0.782367806000783}
			------------EPOCH 8---------------
Loss:  tensor(0.5795, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.7354, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6201, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0262, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1788, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5796, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6221, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5424, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3058, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4071, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1831, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3568, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0354, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1797, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3207, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2676, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3025, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2469, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3794, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4230, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1533, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4243, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.6968, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0711, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.34902353530295444, 'recall': 0.5699100572363042, 'f1': 0.4329192546583851, 'number': 1223}, 'P': {'precision': 0.6042707860063607, 'recall': 0.6226591760299626, 'f1': 0.6133271846898779, 'number': 2136}, 'overall_precision': 0.48284897570271557, 'overall_recall': 0.6034534087526049, 'overall_f1': 0.5364562657139077, 'overall_accuracy': 0.7909335737759375}
			------------EPOCH 9---------------
Loss:  tensor(0.2119, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4305, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2558, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0114, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0250, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1219, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3172, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3281, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3325, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1684, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2411, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1557, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2644, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0192, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2209, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3591, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1912, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2501, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1686, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1956, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1479, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0534, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1377, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4247, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0505, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3692390139335477, 'recall': 0.5633687653311529, 'f1': 0.44609906118484943, 'number': 1223}, 'P': {'precision': 0.5944022770398482, 'recall': 0.5866104868913857, 'f1': 0.5904806786050896, 'number': 2136}, 'overall_precision': 0.4886763965777554, 'overall_recall': 0.5781482584102411, 'overall_f1': 0.5296604391108686, 'overall_accuracy': 0.7947181720034643}
			------------EPOCH 10---------------
Loss:  tensor(0.1479, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2168, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0191, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0166, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1492, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4465, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3414, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3431, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1452, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1224, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1204, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1299, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0065, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1084, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1137, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1342, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1873, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1873, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1823, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1035, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1885, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.5242, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0253, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.38540245566166437, 'recall': 0.46197874080130824, 'f1': 0.4202305689847527, 'number': 1223}, 'P': {'precision': 0.5652680652680653, 'recall': 0.6811797752808989, 'f1': 0.6178343949044587, 'number': 2136}, 'overall_precision': 0.5, 'overall_recall': 0.6013694551949985, 'overall_f1': 0.5460197323962697, 'overall_accuracy': 0.8095599663063983}
			------------EPOCH 11---------------
Loss:  tensor(0.1390, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2286, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1146, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0029, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0081, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1000, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1904, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1694, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3084, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1528, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2200, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1184, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1248, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0079, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1224, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1050, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0758, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1069, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1696, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0884, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0749, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0461, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1485, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4848, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0356, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3790199081163859, 'recall': 0.40474243663123466, 'f1': 0.3914590747330961, 'number': 1223}, 'P': {'precision': 0.5727824109173616, 'recall': 0.7073970037453183, 'f1': 0.6330121491411814, 'number': 2136}, 'overall_precision': 0.5086206896551724, 'overall_recall': 0.5972015480797856, 'overall_f1': 0.5493632753662877, 'overall_accuracy': 0.7974706070780292}
			------------EPOCH 12---------------
Loss:  tensor(0.1699, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2516, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1358, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0110, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0867, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1387, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0909, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0671, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0605, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0775, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0855, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0062, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0729, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0531, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1350, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0936, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1397, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0824, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0965, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0627, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1025, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.4012, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0282, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4654605263157895, 'recall': 0.4627964022894522, 'f1': 0.46412464124641245, 'number': 1223}, 'P': {'precision': 0.5860428231562252, 'recall': 0.6919475655430711, 'f1': 0.634607127522542, 'number': 2136}, 'overall_precision': 0.5468164794007491, 'overall_recall': 0.6085144388210777, 'overall_f1': 0.5760180357897703, 'overall_accuracy': 0.8112327824508536}
			------------EPOCH 13---------------
Loss:  tensor(0.0786, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1620, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0963, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0063, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0861, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1116, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0419, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2196, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0653, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0417, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0624, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0535, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0032, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0627, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0304, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0294, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0430, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1203, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0358, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0389, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0221, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0424, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.3271, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0063, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.3832039520112915, 'recall': 0.4439901880621423, 'f1': 0.41136363636363643, 'number': 1223}, 'P': {'precision': 0.5457972107048624, 'recall': 0.6779026217228464, 'f1': 0.6047191480476091, 'number': 2136}, 'overall_precision': 0.4891891891891892, 'overall_recall': 0.5927359333134862, 'overall_f1': 0.5360075380266524, 'overall_accuracy': 0.8050160756445087}
			------------EPOCH 14---------------
Loss:  tensor(0.0540, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1160, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0435, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0013, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0036, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0693, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0643, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0196, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1215, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0575, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0277, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0515, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0338, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0548, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0203, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0183, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0325, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1098, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0223, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0234, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0072, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0306, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2873, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0032, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4162473040977714, 'recall': 0.473426001635323, 'f1': 0.44299923488905896, 'number': 1223}, 'P': {'precision': 0.603265306122449, 'recall': 0.6919475655430711, 'f1': 0.6445704317488008, 'number': 2136}, 'overall_precision': 0.5355376204113512, 'overall_recall': 0.612384638285204, 'overall_f1': 0.571388888888889, 'overall_accuracy': 0.8106395852365077}
			------------EPOCH 15---------------
Loss:  tensor(0.0452, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0749, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0235, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0639, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0469, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0110, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1463, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0307, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0142, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0471, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0223, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0485, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0108, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0119, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0197, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0161, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0171, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0037, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0249, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2573, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0024, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.41296687808315713, 'recall': 0.47914963205233035, 'f1': 0.44360333080999237, 'number': 1223}, 'P': {'precision': 0.587288817377313, 'recall': 0.6835205992509363, 'f1': 0.6317611423626136, 'number': 2136}, 'overall_precision': 0.523943661971831, 'overall_recall': 0.6091098541232509, 'overall_f1': 0.5633259911894273, 'overall_accuracy': 0.8105565376264993}
			------------EPOCH 16---------------
Loss:  tensor(0.0429, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0496, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0148, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0605, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0307, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0078, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1112, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0948, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0102, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0439, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0135, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0449, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0079, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0089, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0156, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0905, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0087, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0135, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0218, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2349, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0019, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.425019425019425, 'recall': 0.4472608340147179, 'f1': 0.4358565737051793, 'number': 1223}, 'P': {'precision': 0.5840776699029127, 'recall': 0.704119850187266, 'f1': 0.6385056251326682, 'number': 2136}, 'overall_precision': 0.5310719834282756, 'overall_recall': 0.6105983923786842, 'overall_f1': 0.5680653649079076, 'overall_accuracy': 0.8099514764678665}
			------------EPOCH 17---------------
Loss:  tensor(0.0517, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0281, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0122, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0577, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0206, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0075, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2829, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0109, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0087, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0413, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0099, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0426, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0065, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0072, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0129, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0864, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0068, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0125, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0206, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2203, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0016, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.41605839416058393, 'recall': 0.4660670482420278, 'f1': 0.43964519861164675, 'number': 1223}, 'P': {'precision': 0.5975708502024292, 'recall': 0.6910112359550562, 'f1': 0.6409031697785498, 'number': 2136}, 'overall_precision': 0.5328125, 'overall_recall': 0.6091098541232509, 'overall_f1': 0.5684122794832617, 'overall_accuracy': 0.8103785784621955}
			------------EPOCH 18---------------
Loss:  tensor(0.0354, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0179, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0086, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0558, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0150, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0057, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1490, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0251, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0060, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0412, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0088, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0421, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0056, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0059, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0120, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0839, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0055, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0121, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0195, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2138, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0014, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4050632911392405, 'recall': 0.47097301717089124, 'f1': 0.435538752362949, 'number': 1223}, 'P': {'precision': 0.5836978131212723, 'recall': 0.6872659176029963, 'f1': 0.631262094173296, 'number': 2136}, 'overall_precision': 0.5191770383540767, 'overall_recall': 0.6085144388210777, 'overall_f1': 0.5603070175438596, 'overall_accuracy': 0.8124784966009799}
			------------EPOCH 19---------------
Loss:  tensor(0.0324, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0129, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0071, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0579, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0154, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0047, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0985, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0419, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0053, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0423, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0082, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0415, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0049, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0056, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0093, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0816, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0043, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0110, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0014, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0170, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1942, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0012, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.42529585798816566, 'recall': 0.47015535568274736, 'f1': 0.44660194174757284, 'number': 1223}, 'P': {'precision': 0.5955465587044534, 'recall': 0.6886704119850188, 'f1': 0.6387320885801129, 'number': 2136}, 'overall_precision': 0.5353218210361067, 'overall_recall': 0.6091098541232509, 'overall_f1': 0.5698370700459545, 'overall_accuracy': 0.8105921294593601}
			------------EPOCH 20---------------
Loss:  tensor(0.0306, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0101, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0052, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0571, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0198, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0040, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1039, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0263, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0041, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0450, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0072, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0390, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0057, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0088, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0825, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0033, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0108, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0011, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0147, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2063, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0010, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.41302681992337165, 'recall': 0.44071954210956665, 'f1': 0.4264240506329114, 'number': 1223}, 'P': {'precision': 0.5635130041462495, 'recall': 0.6999063670411985, 'f1': 0.6243474629358947, 'number': 2136}, 'overall_precision': 0.5138959070237493, 'overall_recall': 0.6055373623102114, 'overall_f1': 0.5559655596555965, 'overall_accuracy': 0.8111141430079845}
			------------EPOCH 21---------------
Loss:  tensor(0.0303, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0086, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0091, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0630, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0110, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0034, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0959, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0299, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0483, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0068, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0409, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0036, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0050, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0083, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0819, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0030, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0099, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0010, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0102, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1951, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0008, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.40166782487838776, 'recall': 0.47260834014717906, 'f1': 0.4342599549211119, 'number': 1223}, 'P': {'precision': 0.5956840390879479, 'recall': 0.6849250936329588, 'f1': 0.6371951219512194, 'number': 2136}, 'overall_precision': 0.52400513478819, 'overall_recall': 0.6076213158678178, 'overall_f1': 0.5627240143369174, 'overall_accuracy': 0.8103311226850479}
			------------EPOCH 22---------------
Loss:  tensor(0.0425, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0076, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0077, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0576, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0091, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0030, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0911, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0421, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0033, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0499, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0063, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0356, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0031, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0032, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0126, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0732, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0026, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0087, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0060, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1979, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0007, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.40456890198968315, 'recall': 0.4488961569910057, 'f1': 0.4255813953488372, 'number': 1223}, 'P': {'precision': 0.5762842796446505, 'recall': 0.6985018726591761, 'f1': 0.6315343915343915, 'number': 2136}, 'overall_precision': 0.5172326406487583, 'overall_recall': 0.6076213158678178, 'overall_f1': 0.5587953456536618, 'overall_accuracy': 0.8098209730807104}
			------------EPOCH 23---------------
Loss:  tensor(0.0329, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0062, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0080, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0446, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0082, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0029, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1337, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0179, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0032, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0399, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0062, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0402, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0031, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0030, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0166, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0662, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0025, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0097, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0009, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1938, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0007, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.43207126948775054, 'recall': 0.4758789860997547, 'f1': 0.4529182879377432, 'number': 1223}, 'P': {'precision': 0.5782046256370051, 'recall': 0.6905430711610487, 'f1': 0.629400469383401, 'number': 2136}, 'overall_precision': 0.5277065161621344, 'overall_recall': 0.612384638285204, 'overall_f1': 0.5669009232465205, 'overall_accuracy': 0.8127869591524398}
			------------EPOCH 24---------------
Loss:  tensor(0.0386, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0056, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0066, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0438, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0074, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0026, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0923, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0527, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0030, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0396, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0059, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0313, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0027, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0028, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0088, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0581, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0077, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2052, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0006, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4194018964259664, 'recall': 0.47015535568274736, 'f1': 0.4433307632999228, 'number': 1223}, 'P': {'precision': 0.5900793650793651, 'recall': 0.6961610486891385, 'f1': 0.6387457044673539, 'number': 2136}, 'overall_precision': 0.52994088923156, 'overall_recall': 0.6138731765406371, 'overall_f1': 0.5688275862068967, 'overall_accuracy': 0.8103429866293348}
			------------EPOCH 25---------------
Loss:  tensor(0.0252, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0064, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0043, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0431, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0086, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0031, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1444, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0135, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0030, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0369, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0063, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0499, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0032, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0030, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0082, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0661, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0097, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0008, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.2119, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0006, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.44480257856567285, 'recall': 0.4513491414554375, 'f1': 0.44805194805194803, 'number': 1223}, 'P': {'precision': 0.5586053412462908, 'recall': 0.7050561797752809, 'f1': 0.6233443708609272, 'number': 2136}, 'overall_precision': 0.5227330454660909, 'overall_recall': 0.6126823459362906, 'overall_f1': 0.5641447368421052, 'overall_accuracy': 0.8127039115424314}
			------------EPOCH 26---------------
Loss:  tensor(0.0255, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0041, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0033, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0409, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0069, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0917, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0514, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0029, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0371, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0056, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0262, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0058, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0028, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0064, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0483, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0067, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1469, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0005, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.43131462333825704, 'recall': 0.47751430907604253, 'f1': 0.4532402017850214, 'number': 1223}, 'P': {'precision': 0.5806577916992952, 'recall': 0.6942883895131086, 'f1': 0.632409381663113, 'number': 2136}, 'overall_precision': 0.5289150460593655, 'overall_recall': 0.6153617147960703, 'overall_f1': 0.5688729874776387, 'overall_accuracy': 0.8109955035651153}
			------------EPOCH 27---------------
Loss:  tensor(0.0338, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0035, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0025, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0354, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0069, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0025, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1166, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0246, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0338, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0054, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0205, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0024, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0030, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0056, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0411, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0033, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0007, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1424, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0005, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4688026981450253, 'recall': 0.45461978740801307, 'f1': 0.4616023246160232, 'number': 1223}, 'P': {'precision': 0.5726235741444867, 'recall': 0.7050561797752809, 'f1': 0.6319765002098195, 'number': 2136}, 'overall_precision': 0.540356394129979, 'overall_recall': 0.6138731765406371, 'overall_f1': 0.5747735191637631, 'overall_accuracy': 0.8149699249012327}
			------------EPOCH 28---------------
Loss:  tensor(0.0425, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0028, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0023, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0371, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0050, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1037, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0225, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0351, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0049, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0219, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0050, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0401, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1375, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0005, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.39918533604887985, 'recall': 0.4807849550286182, 'f1': 0.4362017804154303, 'number': 1223}, 'P': {'precision': 0.5814234016887817, 'recall': 0.6769662921348315, 'f1': 0.6255678131083713, 'number': 2136}, 'overall_precision': 0.5136363636363637, 'overall_recall': 0.6055373623102114, 'overall_f1': 0.5558136357425878, 'overall_accuracy': 0.8101057077435965}
			------------EPOCH 29---------------
Loss:  tensor(0.0451, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0025, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0021, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0404, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0075, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0857, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0405, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0403, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0048, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0002, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0225, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0050, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0446, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0053, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0006, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0017, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1281, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0005, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4407467532467532, 'recall': 0.4439901880621423, 'f1': 0.4423625254582485, 'number': 1223}, 'P': {'precision': 0.5731150038080731, 'recall': 0.7045880149812734, 'f1': 0.6320873582528349, 'number': 2136}, 'overall_precision': 0.5308449974079834, 'overall_recall': 0.6097052694254242, 'overall_f1': 0.5675488430095609, 'overall_accuracy': 0.8115887007794611}
			------------EPOCH 30---------------
Loss:  tensor(0.0204, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0022, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0001, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0359, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0038, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0019, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1267, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0168, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0427, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0048, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0003, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0283, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0020, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0018, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0065, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0547, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0015, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0053, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0005, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0016, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.1140, device='cuda:0', grad_fn=<DivBackward0>)
Loss:  tensor(0.0004, device='cuda:0', grad_fn=<DivBackward0>)
				 {'C': {'precision': 0.4240177909562639, 'recall': 0.4677023712183156, 'f1': 0.4447900466562986, 'number': 1223}, 'P': {'precision': 0.5723684210526315, 'recall': 0.6924157303370787, 'f1': 0.6266949152542373, 'number': 2136}, 'overall_precision': 0.521484871599288, 'overall_recall': 0.6105983923786842, 'overall_f1': 0.5625342841470105, 'overall_accuracy': 0.8132140611467689}
